Upload folder using huggingface_hub

checkpoint-1000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen2.5-7B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.12.0

checkpoint-1000/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "q_proj",
+    "up_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": true
+}

checkpoint-1000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:595ff0a38d1613cb7f3ba4687df0e4229d9c8b201b4b49f28dfed7fcd088c02e
+size 645976488

checkpoint-1000/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-1000/latest

@@ -0,0 +1 @@
+global_step1000

checkpoint-1000/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-1000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-1000/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-1000/trainer_state.json

@@ -0,0 +1,749 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6027727546714888,
+  "eval_steps": 400,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006027727546714889,
+      "grad_norm": 6.095676898956299,
+      "learning_rate": 1.0040160642570282e-06,
+      "loss": 1.8585,
+      "step": 10
+    },
+    {
+      "epoch": 0.012055455093429777,
+      "grad_norm": 0.6981305480003357,
+      "learning_rate": 2.0080321285140564e-06,
+      "loss": 1.8607,
+      "step": 20
+    },
+    {
+      "epoch": 0.018083182640144666,
+      "grad_norm": 0.33439213037490845,
+      "learning_rate": 3.0120481927710846e-06,
+      "loss": 1.8035,
+      "step": 30
+    },
+    {
+      "epoch": 0.024110910186859555,
+      "grad_norm": 0.2445673942565918,
+      "learning_rate": 4.016064257028113e-06,
+      "loss": 1.7827,
+      "step": 40
+    },
+    {
+      "epoch": 0.030138637733574444,
+      "grad_norm": 0.22277717292308807,
+      "learning_rate": 5.020080321285141e-06,
+      "loss": 1.7306,
+      "step": 50
+    },
+    {
+      "epoch": 0.03616636528028933,
+      "grad_norm": 0.16803288459777832,
+      "learning_rate": 6.024096385542169e-06,
+      "loss": 1.728,
+      "step": 60
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.1401725709438324,
+      "learning_rate": 7.028112449799197e-06,
+      "loss": 1.6623,
+      "step": 70
+    },
+    {
+      "epoch": 0.04822182037371911,
+      "grad_norm": 0.14054332673549652,
+      "learning_rate": 8.032128514056226e-06,
+      "loss": 1.6226,
+      "step": 80
+    },
+    {
+      "epoch": 0.054249547920433995,
+      "grad_norm": 0.1356767863035202,
+      "learning_rate": 9.036144578313253e-06,
+      "loss": 1.6497,
+      "step": 90
+    },
+    {
+      "epoch": 0.06027727546714889,
+      "grad_norm": 0.14919587969779968,
+      "learning_rate": 1.0040160642570281e-05,
+      "loss": 1.6452,
+      "step": 100
+    },
+    {
+      "epoch": 0.06630500301386377,
+      "grad_norm": 0.16640327870845795,
+      "learning_rate": 1.104417670682731e-05,
+      "loss": 1.6311,
+      "step": 110
+    },
+    {
+      "epoch": 0.07233273056057866,
+      "grad_norm": 0.19770000874996185,
+      "learning_rate": 1.2048192771084338e-05,
+      "loss": 1.5925,
+      "step": 120
+    },
+    {
+      "epoch": 0.07836045810729356,
+      "grad_norm": 0.20399077236652374,
+      "learning_rate": 1.3052208835341367e-05,
+      "loss": 1.5825,
+      "step": 130
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.1512237936258316,
+      "learning_rate": 1.4056224899598394e-05,
+      "loss": 1.5478,
+      "step": 140
+    },
+    {
+      "epoch": 0.09041591320072333,
+      "grad_norm": 0.14794760942459106,
+      "learning_rate": 1.5060240963855424e-05,
+      "loss": 1.5252,
+      "step": 150
+    },
+    {
+      "epoch": 0.09644364074743822,
+      "grad_norm": 0.14903859794139862,
+      "learning_rate": 1.606425702811245e-05,
+      "loss": 1.5427,
+      "step": 160
+    },
+    {
+      "epoch": 0.1024713682941531,
+      "grad_norm": 0.1387884020805359,
+      "learning_rate": 1.706827309236948e-05,
+      "loss": 1.554,
+      "step": 170
+    },
+    {
+      "epoch": 0.10849909584086799,
+      "grad_norm": 0.14016373455524445,
+      "learning_rate": 1.8072289156626505e-05,
+      "loss": 1.5415,
+      "step": 180
+    },
+    {
+      "epoch": 0.11452682338758288,
+      "grad_norm": 0.14568917453289032,
+      "learning_rate": 1.9076305220883535e-05,
+      "loss": 1.5516,
+      "step": 190
+    },
+    {
+      "epoch": 0.12055455093429777,
+      "grad_norm": 0.14068633317947388,
+      "learning_rate": 2.0080321285140562e-05,
+      "loss": 1.5437,
+      "step": 200
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.1428183913230896,
+      "learning_rate": 2.1084337349397593e-05,
+      "loss": 1.5315,
+      "step": 210
+    },
+    {
+      "epoch": 0.13261000602772754,
+      "grad_norm": 0.1416633427143097,
+      "learning_rate": 2.208835341365462e-05,
+      "loss": 1.5276,
+      "step": 220
+    },
+    {
+      "epoch": 0.13863773357444242,
+      "grad_norm": 0.14373181760311127,
+      "learning_rate": 2.309236947791165e-05,
+      "loss": 1.5275,
+      "step": 230
+    },
+    {
+      "epoch": 0.14466546112115733,
+      "grad_norm": 0.16374213993549347,
+      "learning_rate": 2.4096385542168677e-05,
+      "loss": 1.5528,
+      "step": 240
+    },
+    {
+      "epoch": 0.1506931886678722,
+      "grad_norm": 0.13819484412670135,
+      "learning_rate": 2.5100401606425704e-05,
+      "loss": 1.5389,
+      "step": 250
+    },
+    {
+      "epoch": 0.1567209162145871,
+      "grad_norm": 0.14697293937206268,
+      "learning_rate": 2.6104417670682734e-05,
+      "loss": 1.5229,
+      "step": 260
+    },
+    {
+      "epoch": 0.162748643761302,
+      "grad_norm": 0.14389023184776306,
+      "learning_rate": 2.7108433734939758e-05,
+      "loss": 1.5376,
+      "step": 270
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.14889676868915558,
+      "learning_rate": 2.8112449799196788e-05,
+      "loss": 1.5508,
+      "step": 280
+    },
+    {
+      "epoch": 0.17480409885473178,
+      "grad_norm": 0.15245158970355988,
+      "learning_rate": 2.911646586345382e-05,
+      "loss": 1.5342,
+      "step": 290
+    },
+    {
+      "epoch": 0.18083182640144665,
+      "grad_norm": 0.143811896443367,
+      "learning_rate": 3.012048192771085e-05,
+      "loss": 1.5123,
+      "step": 300
+    },
+    {
+      "epoch": 0.18685955394816153,
+      "grad_norm": 0.14010299742221832,
+      "learning_rate": 3.112449799196787e-05,
+      "loss": 1.5355,
+      "step": 310
+    },
+    {
+      "epoch": 0.19288728149487644,
+      "grad_norm": 0.14191724359989166,
+      "learning_rate": 3.21285140562249e-05,
+      "loss": 1.5052,
+      "step": 320
+    },
+    {
+      "epoch": 0.19891500904159132,
+      "grad_norm": 0.13444629311561584,
+      "learning_rate": 3.313253012048193e-05,
+      "loss": 1.5357,
+      "step": 330
+    },
+    {
+      "epoch": 0.2049427365883062,
+      "grad_norm": 0.15248367190361023,
+      "learning_rate": 3.413654618473896e-05,
+      "loss": 1.5171,
+      "step": 340
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.16744668781757355,
+      "learning_rate": 3.5140562248995983e-05,
+      "loss": 1.5379,
+      "step": 350
+    },
+    {
+      "epoch": 0.21699819168173598,
+      "grad_norm": 0.1389068365097046,
+      "learning_rate": 3.614457831325301e-05,
+      "loss": 1.5142,
+      "step": 360
+    },
+    {
+      "epoch": 0.2230259192284509,
+      "grad_norm": 0.13487495481967926,
+      "learning_rate": 3.7148594377510044e-05,
+      "loss": 1.5189,
+      "step": 370
+    },
+    {
+      "epoch": 0.22905364677516576,
+      "grad_norm": 0.13311463594436646,
+      "learning_rate": 3.815261044176707e-05,
+      "loss": 1.5166,
+      "step": 380
+    },
+    {
+      "epoch": 0.23508137432188064,
+      "grad_norm": 0.1321990191936493,
+      "learning_rate": 3.91566265060241e-05,
+      "loss": 1.5218,
+      "step": 390
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "grad_norm": 0.1350693255662918,
+      "learning_rate": 4.0160642570281125e-05,
+      "loss": 1.54,
+      "step": 400
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "eval_loss": 1.4954419136047363,
+      "eval_runtime": 110.0553,
+      "eval_samples_per_second": 9.75,
+      "eval_steps_per_second": 2.444,
+      "step": 400
+    },
+    {
+      "epoch": 0.24713682941531043,
+      "grad_norm": 0.1386946737766266,
+      "learning_rate": 4.116465863453816e-05,
+      "loss": 1.5307,
+      "step": 410
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.1321861445903778,
+      "learning_rate": 4.2168674698795186e-05,
+      "loss": 1.5036,
+      "step": 420
+    },
+    {
+      "epoch": 0.2591922845087402,
+      "grad_norm": 0.13090862333774567,
+      "learning_rate": 4.317269076305221e-05,
+      "loss": 1.5036,
+      "step": 430
+    },
+    {
+      "epoch": 0.2652200120554551,
+      "grad_norm": 0.14390167593955994,
+      "learning_rate": 4.417670682730924e-05,
+      "loss": 1.5245,
+      "step": 440
+    },
+    {
+      "epoch": 0.27124773960216997,
+      "grad_norm": 0.14252203702926636,
+      "learning_rate": 4.5180722891566266e-05,
+      "loss": 1.5361,
+      "step": 450
+    },
+    {
+      "epoch": 0.27727546714888485,
+      "grad_norm": 0.1380358338356018,
+      "learning_rate": 4.61847389558233e-05,
+      "loss": 1.5379,
+      "step": 460
+    },
+    {
+      "epoch": 0.2833031946955998,
+      "grad_norm": 0.17792685329914093,
+      "learning_rate": 4.718875502008032e-05,
+      "loss": 1.5107,
+      "step": 470
+    },
+    {
+      "epoch": 0.28933092224231466,
+      "grad_norm": 0.1495760679244995,
+      "learning_rate": 4.8192771084337354e-05,
+      "loss": 1.5063,
+      "step": 480
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.12926514446735382,
+      "learning_rate": 4.919678714859438e-05,
+      "loss": 1.5149,
+      "step": 490
+    },
+    {
+      "epoch": 0.3013863773357444,
+      "grad_norm": 0.12767350673675537,
+      "learning_rate": 4.999997540156068e-05,
+      "loss": 1.5052,
+      "step": 500
+    },
+    {
+      "epoch": 0.3074141048824593,
+      "grad_norm": 0.12568402290344238,
+      "learning_rate": 4.999911446126714e-05,
+      "loss": 1.541,
+      "step": 510
+    },
+    {
+      "epoch": 0.3134418324291742,
+      "grad_norm": 0.1353764533996582,
+      "learning_rate": 4.9997023647413776e-05,
+      "loss": 1.5218,
+      "step": 520
+    },
+    {
+      "epoch": 0.3194695599758891,
+      "grad_norm": 2.775493621826172,
+      "learning_rate": 4.999370306286171e-05,
+      "loss": 1.5301,
+      "step": 530
+    },
+    {
+      "epoch": 0.325497287522604,
+      "grad_norm": 0.12400302290916443,
+      "learning_rate": 4.998915287097269e-05,
+      "loss": 1.4685,
+      "step": 540
+    },
+    {
+      "epoch": 0.33152501506931886,
+      "grad_norm": 0.12950482964515686,
+      "learning_rate": 4.998337329560107e-05,
+      "loss": 1.5264,
+      "step": 550
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.12773030996322632,
+      "learning_rate": 4.99763646210828e-05,
+      "loss": 1.4937,
+      "step": 560
+    },
+    {
+      "epoch": 0.3435804701627486,
+      "grad_norm": 0.12648645043373108,
+      "learning_rate": 4.9968127192221436e-05,
+      "loss": 1.5126,
+      "step": 570
+    },
+    {
+      "epoch": 0.34960819770946355,
+      "grad_norm": 0.12337373942136765,
+      "learning_rate": 4.995866141427117e-05,
+      "loss": 1.4816,
+      "step": 580
+    },
+    {
+      "epoch": 0.35563592525617843,
+      "grad_norm": 0.1327151358127594,
+      "learning_rate": 4.99479677529169e-05,
+      "loss": 1.5131,
+      "step": 590
+    },
+    {
+      "epoch": 0.3616636528028933,
+      "grad_norm": 0.13992607593536377,
+      "learning_rate": 4.9936046734251313e-05,
+      "loss": 1.5089,
+      "step": 600
+    },
+    {
+      "epoch": 0.3676913803496082,
+      "grad_norm": 0.12979301810264587,
+      "learning_rate": 4.992289894474901e-05,
+      "loss": 1.5032,
+      "step": 610
+    },
+    {
+      "epoch": 0.37371910789632307,
+      "grad_norm": 0.1280072033405304,
+      "learning_rate": 4.990852503123765e-05,
+      "loss": 1.5012,
+      "step": 620
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.1306256204843521,
+      "learning_rate": 4.989292570086612e-05,
+      "loss": 1.4564,
+      "step": 630
+    },
+    {
+      "epoch": 0.3857745629897529,
+      "grad_norm": 0.13198673725128174,
+      "learning_rate": 4.987610172106978e-05,
+      "loss": 1.5106,
+      "step": 640
+    },
+    {
+      "epoch": 0.39180229053646776,
+      "grad_norm": 0.1281399428844452,
+      "learning_rate": 4.9858053919532646e-05,
+      "loss": 1.4834,
+      "step": 650
+    },
+    {
+      "epoch": 0.39783001808318263,
+      "grad_norm": 0.13263563811779022,
+      "learning_rate": 4.983878318414674e-05,
+      "loss": 1.5208,
+      "step": 660
+    },
+    {
+      "epoch": 0.4038577456298975,
+      "grad_norm": 0.13027793169021606,
+      "learning_rate": 4.981829046296836e-05,
+      "loss": 1.4892,
+      "step": 670
+    },
+    {
+      "epoch": 0.4098854731766124,
+      "grad_norm": 0.13578078150749207,
+      "learning_rate": 4.9796576764171446e-05,
+      "loss": 1.4874,
+      "step": 680
+    },
+    {
+      "epoch": 0.4159132007233273,
+      "grad_norm": 0.13204683363437653,
+      "learning_rate": 4.9773643155998e-05,
+      "loss": 1.4867,
+      "step": 690
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.13477781414985657,
+      "learning_rate": 4.974949076670553e-05,
+      "loss": 1.5159,
+      "step": 700
+    },
+    {
+      "epoch": 0.4279686558167571,
+      "grad_norm": 0.13009533286094666,
+      "learning_rate": 4.972412078451151e-05,
+      "loss": 1.4806,
+      "step": 710
+    },
+    {
+      "epoch": 0.43399638336347196,
+      "grad_norm": 0.1321401745080948,
+      "learning_rate": 4.9697534457534966e-05,
+      "loss": 1.4641,
+      "step": 720
+    },
+    {
+      "epoch": 0.44002411091018684,
+      "grad_norm": 0.13463927805423737,
+      "learning_rate": 4.966973309373506e-05,
+      "loss": 1.4924,
+      "step": 730
+    },
+    {
+      "epoch": 0.4460518384569018,
+      "grad_norm": 0.1337679624557495,
+      "learning_rate": 4.9640718060846723e-05,
+      "loss": 1.4805,
+      "step": 740
+    },
+    {
+      "epoch": 0.45207956600361665,
+      "grad_norm": 0.1317020207643509,
+      "learning_rate": 4.96104907863134e-05,
+      "loss": 1.4678,
+      "step": 750
+    },
+    {
+      "epoch": 0.45810729355033153,
+      "grad_norm": 0.13243788480758667,
+      "learning_rate": 4.957905275721678e-05,
+      "loss": 1.4794,
+      "step": 760
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.13113069534301758,
+      "learning_rate": 4.954640552020369e-05,
+      "loss": 1.4734,
+      "step": 770
+    },
+    {
+      "epoch": 0.4701627486437613,
+      "grad_norm": 0.13361573219299316,
+      "learning_rate": 4.9512550681409946e-05,
+      "loss": 1.4803,
+      "step": 780
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.13844482600688934,
+      "learning_rate": 4.9477489906381414e-05,
+      "loss": 1.4665,
+      "step": 790
+    },
+    {
+      "epoch": 0.4822182037371911,
+      "grad_norm": 0.13865956664085388,
+      "learning_rate": 4.944122491999198e-05,
+      "loss": 1.4872,
+      "step": 800
+    },
+    {
+      "epoch": 0.4822182037371911,
+      "eval_loss": 1.4607902765274048,
+      "eval_runtime": 110.1311,
+      "eval_samples_per_second": 9.743,
+      "eval_steps_per_second": 2.443,
+      "step": 800
+    },
+    {
+      "epoch": 0.488245931283906,
+      "grad_norm": 0.13776348531246185,
+      "learning_rate": 4.9403757506358775e-05,
+      "loss": 1.5016,
+      "step": 810
+    },
+    {
+      "epoch": 0.49427365883062085,
+      "grad_norm": 0.14617718756198883,
+      "learning_rate": 4.936508950875432e-05,
+      "loss": 1.4857,
+      "step": 820
+    },
+    {
+      "epoch": 0.5003013863773358,
+      "grad_norm": 0.14705806970596313,
+      "learning_rate": 4.932522282951594e-05,
+      "loss": 1.4909,
+      "step": 830
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.14169825613498688,
+      "learning_rate": 4.928415942995208e-05,
+      "loss": 1.5106,
+      "step": 840
+    },
+    {
+      "epoch": 0.5123568414707655,
+      "grad_norm": 0.14307105541229248,
+      "learning_rate": 4.924190133024589e-05,
+      "loss": 1.492,
+      "step": 850
+    },
+    {
+      "epoch": 0.5183845690174804,
+      "grad_norm": 0.1347261369228363,
+      "learning_rate": 4.919845060935577e-05,
+      "loss": 1.5063,
+      "step": 860
+    },
+    {
+      "epoch": 0.5244122965641953,
+      "grad_norm": 0.1423402726650238,
+      "learning_rate": 4.915380940491315e-05,
+      "loss": 1.4682,
+      "step": 870
+    },
+    {
+      "epoch": 0.5304400241109102,
+      "grad_norm": 0.14149978756904602,
+      "learning_rate": 4.910797991311733e-05,
+      "loss": 1.4436,
+      "step": 880
+    },
+    {
+      "epoch": 0.5364677516576251,
+      "grad_norm": 0.1491919308900833,
+      "learning_rate": 4.906096438862735e-05,
+      "loss": 1.4851,
+      "step": 890
+    },
+    {
+      "epoch": 0.5424954792043399,
+      "grad_norm": 0.14350004494190216,
+      "learning_rate": 4.9012765144451186e-05,
+      "loss": 1.4546,
+      "step": 900
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.14726224541664124,
+      "learning_rate": 4.896338455183185e-05,
+      "loss": 1.473,
+      "step": 910
+    },
+    {
+      "epoch": 0.5545509342977697,
+      "grad_norm": 0.1460171639919281,
+      "learning_rate": 4.8912825040130814e-05,
+      "loss": 1.4802,
+      "step": 920
+    },
+    {
+      "epoch": 0.5605786618444847,
+      "grad_norm": 0.15167438983917236,
+      "learning_rate": 4.886108909670845e-05,
+      "loss": 1.4542,
+      "step": 930
+    },
+    {
+      "epoch": 0.5666063893911996,
+      "grad_norm": 0.15603500604629517,
+      "learning_rate": 4.880817926680167e-05,
+      "loss": 1.4565,
+      "step": 940
+    },
+    {
+      "epoch": 0.5726341169379144,
+      "grad_norm": 0.14839062094688416,
+      "learning_rate": 4.8754098153398706e-05,
+      "loss": 1.4833,
+      "step": 950
+    },
+    {
+      "epoch": 0.5786618444846293,
+      "grad_norm": 0.14976999163627625,
+      "learning_rate": 4.869884841711107e-05,
+      "loss": 1.4957,
+      "step": 960
+    },
+    {
+      "epoch": 0.5846895720313442,
+      "grad_norm": 0.15414153039455414,
+      "learning_rate": 4.8642432776042616e-05,
+      "loss": 1.4816,
+      "step": 970
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.15507987141609192,
+      "learning_rate": 4.858485400565589e-05,
+      "loss": 1.4728,
+      "step": 980
+    },
+    {
+      "epoch": 0.596745027124774,
+      "grad_norm": 0.1561218947172165,
+      "learning_rate": 4.85261149386355e-05,
+      "loss": 1.4936,
+      "step": 990
+    },
+    {
+      "epoch": 0.6027727546714888,
+      "grad_norm": 0.15994535386562347,
+      "learning_rate": 4.846621846474885e-05,
+      "loss": 1.4705,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 4977,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.81454820016128e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:276772c4ef2f26970b87a5557f24d7c42bdfe56a8d7f3d49ad7c2ec614711f92
+size 7032

checkpoint-1000/zero_to_fp32.py

@@ -0,0 +1,674 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import json
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering Sharded Weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        state_dict_split = split_torch_state_dict_into_shards(state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors}
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard, output_path)
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

checkpoint-1500/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen2.5-7B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.12.0

checkpoint-1500/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "q_proj",
+    "up_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": true
+}

checkpoint-1500/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94314d88f86fb8de9b54f5c5c819c6dfc22dc2bd60b3bfebdf17675266034eba
+size 645976488

checkpoint-1500/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-1500/latest

@@ -0,0 +1 @@
+global_step1500

checkpoint-1500/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-1500/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-1500/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-1500/trainer_state.json

@@ -0,0 +1,1107 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9041591320072333,
+  "eval_steps": 400,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006027727546714889,
+      "grad_norm": 6.095676898956299,
+      "learning_rate": 1.0040160642570282e-06,
+      "loss": 1.8585,
+      "step": 10
+    },
+    {
+      "epoch": 0.012055455093429777,
+      "grad_norm": 0.6981305480003357,
+      "learning_rate": 2.0080321285140564e-06,
+      "loss": 1.8607,
+      "step": 20
+    },
+    {
+      "epoch": 0.018083182640144666,
+      "grad_norm": 0.33439213037490845,
+      "learning_rate": 3.0120481927710846e-06,
+      "loss": 1.8035,
+      "step": 30
+    },
+    {
+      "epoch": 0.024110910186859555,
+      "grad_norm": 0.2445673942565918,
+      "learning_rate": 4.016064257028113e-06,
+      "loss": 1.7827,
+      "step": 40
+    },
+    {
+      "epoch": 0.030138637733574444,
+      "grad_norm": 0.22277717292308807,
+      "learning_rate": 5.020080321285141e-06,
+      "loss": 1.7306,
+      "step": 50
+    },
+    {
+      "epoch": 0.03616636528028933,
+      "grad_norm": 0.16803288459777832,
+      "learning_rate": 6.024096385542169e-06,
+      "loss": 1.728,
+      "step": 60
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.1401725709438324,
+      "learning_rate": 7.028112449799197e-06,
+      "loss": 1.6623,
+      "step": 70
+    },
+    {
+      "epoch": 0.04822182037371911,
+      "grad_norm": 0.14054332673549652,
+      "learning_rate": 8.032128514056226e-06,
+      "loss": 1.6226,
+      "step": 80
+    },
+    {
+      "epoch": 0.054249547920433995,
+      "grad_norm": 0.1356767863035202,
+      "learning_rate": 9.036144578313253e-06,
+      "loss": 1.6497,
+      "step": 90
+    },
+    {
+      "epoch": 0.06027727546714889,
+      "grad_norm": 0.14919587969779968,
+      "learning_rate": 1.0040160642570281e-05,
+      "loss": 1.6452,
+      "step": 100
+    },
+    {
+      "epoch": 0.06630500301386377,
+      "grad_norm": 0.16640327870845795,
+      "learning_rate": 1.104417670682731e-05,
+      "loss": 1.6311,
+      "step": 110
+    },
+    {
+      "epoch": 0.07233273056057866,
+      "grad_norm": 0.19770000874996185,
+      "learning_rate": 1.2048192771084338e-05,
+      "loss": 1.5925,
+      "step": 120
+    },
+    {
+      "epoch": 0.07836045810729356,
+      "grad_norm": 0.20399077236652374,
+      "learning_rate": 1.3052208835341367e-05,
+      "loss": 1.5825,
+      "step": 130
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.1512237936258316,
+      "learning_rate": 1.4056224899598394e-05,
+      "loss": 1.5478,
+      "step": 140
+    },
+    {
+      "epoch": 0.09041591320072333,
+      "grad_norm": 0.14794760942459106,
+      "learning_rate": 1.5060240963855424e-05,
+      "loss": 1.5252,
+      "step": 150
+    },
+    {
+      "epoch": 0.09644364074743822,
+      "grad_norm": 0.14903859794139862,
+      "learning_rate": 1.606425702811245e-05,
+      "loss": 1.5427,
+      "step": 160
+    },
+    {
+      "epoch": 0.1024713682941531,
+      "grad_norm": 0.1387884020805359,
+      "learning_rate": 1.706827309236948e-05,
+      "loss": 1.554,
+      "step": 170
+    },
+    {
+      "epoch": 0.10849909584086799,
+      "grad_norm": 0.14016373455524445,
+      "learning_rate": 1.8072289156626505e-05,
+      "loss": 1.5415,
+      "step": 180
+    },
+    {
+      "epoch": 0.11452682338758288,
+      "grad_norm": 0.14568917453289032,
+      "learning_rate": 1.9076305220883535e-05,
+      "loss": 1.5516,
+      "step": 190
+    },
+    {
+      "epoch": 0.12055455093429777,
+      "grad_norm": 0.14068633317947388,
+      "learning_rate": 2.0080321285140562e-05,
+      "loss": 1.5437,
+      "step": 200
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.1428183913230896,
+      "learning_rate": 2.1084337349397593e-05,
+      "loss": 1.5315,
+      "step": 210
+    },
+    {
+      "epoch": 0.13261000602772754,
+      "grad_norm": 0.1416633427143097,
+      "learning_rate": 2.208835341365462e-05,
+      "loss": 1.5276,
+      "step": 220
+    },
+    {
+      "epoch": 0.13863773357444242,
+      "grad_norm": 0.14373181760311127,
+      "learning_rate": 2.309236947791165e-05,
+      "loss": 1.5275,
+      "step": 230
+    },
+    {
+      "epoch": 0.14466546112115733,
+      "grad_norm": 0.16374213993549347,
+      "learning_rate": 2.4096385542168677e-05,
+      "loss": 1.5528,
+      "step": 240
+    },
+    {
+      "epoch": 0.1506931886678722,
+      "grad_norm": 0.13819484412670135,
+      "learning_rate": 2.5100401606425704e-05,
+      "loss": 1.5389,
+      "step": 250
+    },
+    {
+      "epoch": 0.1567209162145871,
+      "grad_norm": 0.14697293937206268,
+      "learning_rate": 2.6104417670682734e-05,
+      "loss": 1.5229,
+      "step": 260
+    },
+    {
+      "epoch": 0.162748643761302,
+      "grad_norm": 0.14389023184776306,
+      "learning_rate": 2.7108433734939758e-05,
+      "loss": 1.5376,
+      "step": 270
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.14889676868915558,
+      "learning_rate": 2.8112449799196788e-05,
+      "loss": 1.5508,
+      "step": 280
+    },
+    {
+      "epoch": 0.17480409885473178,
+      "grad_norm": 0.15245158970355988,
+      "learning_rate": 2.911646586345382e-05,
+      "loss": 1.5342,
+      "step": 290
+    },
+    {
+      "epoch": 0.18083182640144665,
+      "grad_norm": 0.143811896443367,
+      "learning_rate": 3.012048192771085e-05,
+      "loss": 1.5123,
+      "step": 300
+    },
+    {
+      "epoch": 0.18685955394816153,
+      "grad_norm": 0.14010299742221832,
+      "learning_rate": 3.112449799196787e-05,
+      "loss": 1.5355,
+      "step": 310
+    },
+    {
+      "epoch": 0.19288728149487644,
+      "grad_norm": 0.14191724359989166,
+      "learning_rate": 3.21285140562249e-05,
+      "loss": 1.5052,
+      "step": 320
+    },
+    {
+      "epoch": 0.19891500904159132,
+      "grad_norm": 0.13444629311561584,
+      "learning_rate": 3.313253012048193e-05,
+      "loss": 1.5357,
+      "step": 330
+    },
+    {
+      "epoch": 0.2049427365883062,
+      "grad_norm": 0.15248367190361023,
+      "learning_rate": 3.413654618473896e-05,
+      "loss": 1.5171,
+      "step": 340
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.16744668781757355,
+      "learning_rate": 3.5140562248995983e-05,
+      "loss": 1.5379,
+      "step": 350
+    },
+    {
+      "epoch": 0.21699819168173598,
+      "grad_norm": 0.1389068365097046,
+      "learning_rate": 3.614457831325301e-05,
+      "loss": 1.5142,
+      "step": 360
+    },
+    {
+      "epoch": 0.2230259192284509,
+      "grad_norm": 0.13487495481967926,
+      "learning_rate": 3.7148594377510044e-05,
+      "loss": 1.5189,
+      "step": 370
+    },
+    {
+      "epoch": 0.22905364677516576,
+      "grad_norm": 0.13311463594436646,
+      "learning_rate": 3.815261044176707e-05,
+      "loss": 1.5166,
+      "step": 380
+    },
+    {
+      "epoch": 0.23508137432188064,
+      "grad_norm": 0.1321990191936493,
+      "learning_rate": 3.91566265060241e-05,
+      "loss": 1.5218,
+      "step": 390
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "grad_norm": 0.1350693255662918,
+      "learning_rate": 4.0160642570281125e-05,
+      "loss": 1.54,
+      "step": 400
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "eval_loss": 1.4954419136047363,
+      "eval_runtime": 110.0553,
+      "eval_samples_per_second": 9.75,
+      "eval_steps_per_second": 2.444,
+      "step": 400
+    },
+    {
+      "epoch": 0.24713682941531043,
+      "grad_norm": 0.1386946737766266,
+      "learning_rate": 4.116465863453816e-05,
+      "loss": 1.5307,
+      "step": 410
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.1321861445903778,
+      "learning_rate": 4.2168674698795186e-05,
+      "loss": 1.5036,
+      "step": 420
+    },
+    {
+      "epoch": 0.2591922845087402,
+      "grad_norm": 0.13090862333774567,
+      "learning_rate": 4.317269076305221e-05,
+      "loss": 1.5036,
+      "step": 430
+    },
+    {
+      "epoch": 0.2652200120554551,
+      "grad_norm": 0.14390167593955994,
+      "learning_rate": 4.417670682730924e-05,
+      "loss": 1.5245,
+      "step": 440
+    },
+    {
+      "epoch": 0.27124773960216997,
+      "grad_norm": 0.14252203702926636,
+      "learning_rate": 4.5180722891566266e-05,
+      "loss": 1.5361,
+      "step": 450
+    },
+    {
+      "epoch": 0.27727546714888485,
+      "grad_norm": 0.1380358338356018,
+      "learning_rate": 4.61847389558233e-05,
+      "loss": 1.5379,
+      "step": 460
+    },
+    {
+      "epoch": 0.2833031946955998,
+      "grad_norm": 0.17792685329914093,
+      "learning_rate": 4.718875502008032e-05,
+      "loss": 1.5107,
+      "step": 470
+    },
+    {
+      "epoch": 0.28933092224231466,
+      "grad_norm": 0.1495760679244995,
+      "learning_rate": 4.8192771084337354e-05,
+      "loss": 1.5063,
+      "step": 480
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.12926514446735382,
+      "learning_rate": 4.919678714859438e-05,
+      "loss": 1.5149,
+      "step": 490
+    },
+    {
+      "epoch": 0.3013863773357444,
+      "grad_norm": 0.12767350673675537,
+      "learning_rate": 4.999997540156068e-05,
+      "loss": 1.5052,
+      "step": 500
+    },
+    {
+      "epoch": 0.3074141048824593,
+      "grad_norm": 0.12568402290344238,
+      "learning_rate": 4.999911446126714e-05,
+      "loss": 1.541,
+      "step": 510
+    },
+    {
+      "epoch": 0.3134418324291742,
+      "grad_norm": 0.1353764533996582,
+      "learning_rate": 4.9997023647413776e-05,
+      "loss": 1.5218,
+      "step": 520
+    },
+    {
+      "epoch": 0.3194695599758891,
+      "grad_norm": 2.775493621826172,
+      "learning_rate": 4.999370306286171e-05,
+      "loss": 1.5301,
+      "step": 530
+    },
+    {
+      "epoch": 0.325497287522604,
+      "grad_norm": 0.12400302290916443,
+      "learning_rate": 4.998915287097269e-05,
+      "loss": 1.4685,
+      "step": 540
+    },
+    {
+      "epoch": 0.33152501506931886,
+      "grad_norm": 0.12950482964515686,
+      "learning_rate": 4.998337329560107e-05,
+      "loss": 1.5264,
+      "step": 550
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.12773030996322632,
+      "learning_rate": 4.99763646210828e-05,
+      "loss": 1.4937,
+      "step": 560
+    },
+    {
+      "epoch": 0.3435804701627486,
+      "grad_norm": 0.12648645043373108,
+      "learning_rate": 4.9968127192221436e-05,
+      "loss": 1.5126,
+      "step": 570
+    },
+    {
+      "epoch": 0.34960819770946355,
+      "grad_norm": 0.12337373942136765,
+      "learning_rate": 4.995866141427117e-05,
+      "loss": 1.4816,
+      "step": 580
+    },
+    {
+      "epoch": 0.35563592525617843,
+      "grad_norm": 0.1327151358127594,
+      "learning_rate": 4.99479677529169e-05,
+      "loss": 1.5131,
+      "step": 590
+    },
+    {
+      "epoch": 0.3616636528028933,
+      "grad_norm": 0.13992607593536377,
+      "learning_rate": 4.9936046734251313e-05,
+      "loss": 1.5089,
+      "step": 600
+    },
+    {
+      "epoch": 0.3676913803496082,
+      "grad_norm": 0.12979301810264587,
+      "learning_rate": 4.992289894474901e-05,
+      "loss": 1.5032,
+      "step": 610
+    },
+    {
+      "epoch": 0.37371910789632307,
+      "grad_norm": 0.1280072033405304,
+      "learning_rate": 4.990852503123765e-05,
+      "loss": 1.5012,
+      "step": 620
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.1306256204843521,
+      "learning_rate": 4.989292570086612e-05,
+      "loss": 1.4564,
+      "step": 630
+    },
+    {
+      "epoch": 0.3857745629897529,
+      "grad_norm": 0.13198673725128174,
+      "learning_rate": 4.987610172106978e-05,
+      "loss": 1.5106,
+      "step": 640
+    },
+    {
+      "epoch": 0.39180229053646776,
+      "grad_norm": 0.1281399428844452,
+      "learning_rate": 4.9858053919532646e-05,
+      "loss": 1.4834,
+      "step": 650
+    },
+    {
+      "epoch": 0.39783001808318263,
+      "grad_norm": 0.13263563811779022,
+      "learning_rate": 4.983878318414674e-05,
+      "loss": 1.5208,
+      "step": 660
+    },
+    {
+      "epoch": 0.4038577456298975,
+      "grad_norm": 0.13027793169021606,
+      "learning_rate": 4.981829046296836e-05,
+      "loss": 1.4892,
+      "step": 670
+    },
+    {
+      "epoch": 0.4098854731766124,
+      "grad_norm": 0.13578078150749207,
+      "learning_rate": 4.9796576764171446e-05,
+      "loss": 1.4874,
+      "step": 680
+    },
+    {
+      "epoch": 0.4159132007233273,
+      "grad_norm": 0.13204683363437653,
+      "learning_rate": 4.9773643155998e-05,
+      "loss": 1.4867,
+      "step": 690
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.13477781414985657,
+      "learning_rate": 4.974949076670553e-05,
+      "loss": 1.5159,
+      "step": 700
+    },
+    {
+      "epoch": 0.4279686558167571,
+      "grad_norm": 0.13009533286094666,
+      "learning_rate": 4.972412078451151e-05,
+      "loss": 1.4806,
+      "step": 710
+    },
+    {
+      "epoch": 0.43399638336347196,
+      "grad_norm": 0.1321401745080948,
+      "learning_rate": 4.9697534457534966e-05,
+      "loss": 1.4641,
+      "step": 720
+    },
+    {
+      "epoch": 0.44002411091018684,
+      "grad_norm": 0.13463927805423737,
+      "learning_rate": 4.966973309373506e-05,
+      "loss": 1.4924,
+      "step": 730
+    },
+    {
+      "epoch": 0.4460518384569018,
+      "grad_norm": 0.1337679624557495,
+      "learning_rate": 4.9640718060846723e-05,
+      "loss": 1.4805,
+      "step": 740
+    },
+    {
+      "epoch": 0.45207956600361665,
+      "grad_norm": 0.1317020207643509,
+      "learning_rate": 4.96104907863134e-05,
+      "loss": 1.4678,
+      "step": 750
+    },
+    {
+      "epoch": 0.45810729355033153,
+      "grad_norm": 0.13243788480758667,
+      "learning_rate": 4.957905275721678e-05,
+      "loss": 1.4794,
+      "step": 760
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.13113069534301758,
+      "learning_rate": 4.954640552020369e-05,
+      "loss": 1.4734,
+      "step": 770
+    },
+    {
+      "epoch": 0.4701627486437613,
+      "grad_norm": 0.13361573219299316,
+      "learning_rate": 4.9512550681409946e-05,
+      "loss": 1.4803,
+      "step": 780
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.13844482600688934,
+      "learning_rate": 4.9477489906381414e-05,
+      "loss": 1.4665,
+      "step": 790
+    },
+    {
+      "epoch": 0.4822182037371911,
+      "grad_norm": 0.13865956664085388,
+      "learning_rate": 4.944122491999198e-05,
+      "loss": 1.4872,
+      "step": 800
+    },
+    {
+      "epoch": 0.4822182037371911,
+      "eval_loss": 1.4607902765274048,
+      "eval_runtime": 110.1311,
+      "eval_samples_per_second": 9.743,
+      "eval_steps_per_second": 2.443,
+      "step": 800
+    },
+    {
+      "epoch": 0.488245931283906,
+      "grad_norm": 0.13776348531246185,
+      "learning_rate": 4.9403757506358775e-05,
+      "loss": 1.5016,
+      "step": 810
+    },
+    {
+      "epoch": 0.49427365883062085,
+      "grad_norm": 0.14617718756198883,
+      "learning_rate": 4.936508950875432e-05,
+      "loss": 1.4857,
+      "step": 820
+    },
+    {
+      "epoch": 0.5003013863773358,
+      "grad_norm": 0.14705806970596313,
+      "learning_rate": 4.932522282951594e-05,
+      "loss": 1.4909,
+      "step": 830
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.14169825613498688,
+      "learning_rate": 4.928415942995208e-05,
+      "loss": 1.5106,
+      "step": 840
+    },
+    {
+      "epoch": 0.5123568414707655,
+      "grad_norm": 0.14307105541229248,
+      "learning_rate": 4.924190133024589e-05,
+      "loss": 1.492,
+      "step": 850
+    },
+    {
+      "epoch": 0.5183845690174804,
+      "grad_norm": 0.1347261369228363,
+      "learning_rate": 4.919845060935577e-05,
+      "loss": 1.5063,
+      "step": 860
+    },
+    {
+      "epoch": 0.5244122965641953,
+      "grad_norm": 0.1423402726650238,
+      "learning_rate": 4.915380940491315e-05,
+      "loss": 1.4682,
+      "step": 870
+    },
+    {
+      "epoch": 0.5304400241109102,
+      "grad_norm": 0.14149978756904602,
+      "learning_rate": 4.910797991311733e-05,
+      "loss": 1.4436,
+      "step": 880
+    },
+    {
+      "epoch": 0.5364677516576251,
+      "grad_norm": 0.1491919308900833,
+      "learning_rate": 4.906096438862735e-05,
+      "loss": 1.4851,
+      "step": 890
+    },
+    {
+      "epoch": 0.5424954792043399,
+      "grad_norm": 0.14350004494190216,
+      "learning_rate": 4.9012765144451186e-05,
+      "loss": 1.4546,
+      "step": 900
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.14726224541664124,
+      "learning_rate": 4.896338455183185e-05,
+      "loss": 1.473,
+      "step": 910
+    },
+    {
+      "epoch": 0.5545509342977697,
+      "grad_norm": 0.1460171639919281,
+      "learning_rate": 4.8912825040130814e-05,
+      "loss": 1.4802,
+      "step": 920
+    },
+    {
+      "epoch": 0.5605786618444847,
+      "grad_norm": 0.15167438983917236,
+      "learning_rate": 4.886108909670845e-05,
+      "loss": 1.4542,
+      "step": 930
+    },
+    {
+      "epoch": 0.5666063893911996,
+      "grad_norm": 0.15603500604629517,
+      "learning_rate": 4.880817926680167e-05,
+      "loss": 1.4565,
+      "step": 940
+    },
+    {
+      "epoch": 0.5726341169379144,
+      "grad_norm": 0.14839062094688416,
+      "learning_rate": 4.8754098153398706e-05,
+      "loss": 1.4833,
+      "step": 950
+    },
+    {
+      "epoch": 0.5786618444846293,
+      "grad_norm": 0.14976999163627625,
+      "learning_rate": 4.869884841711107e-05,
+      "loss": 1.4957,
+      "step": 960
+    },
+    {
+      "epoch": 0.5846895720313442,
+      "grad_norm": 0.15414153039455414,
+      "learning_rate": 4.8642432776042616e-05,
+      "loss": 1.4816,
+      "step": 970
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.15507987141609192,
+      "learning_rate": 4.858485400565589e-05,
+      "loss": 1.4728,
+      "step": 980
+    },
+    {
+      "epoch": 0.596745027124774,
+      "grad_norm": 0.1561218947172165,
+      "learning_rate": 4.85261149386355e-05,
+      "loss": 1.4936,
+      "step": 990
+    },
+    {
+      "epoch": 0.6027727546714888,
+      "grad_norm": 0.15994535386562347,
+      "learning_rate": 4.846621846474885e-05,
+      "loss": 1.4705,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6088004822182037,
+      "grad_norm": 0.14968565106391907,
+      "learning_rate": 4.840516753070387e-05,
+      "loss": 1.4638,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6148282097649186,
+      "grad_norm": 0.1536279022693634,
+      "learning_rate": 4.834296514000416e-05,
+      "loss": 1.4757,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6208559373116335,
+      "grad_norm": 0.1513814926147461,
+      "learning_rate": 4.827961435280112e-05,
+      "loss": 1.4548,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6268836648583485,
+      "grad_norm": 0.16222743690013885,
+      "learning_rate": 4.821511828574349e-05,
+      "loss": 1.4643,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.1569400429725647,
+      "learning_rate": 4.814948011182395e-05,
+      "loss": 1.4459,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6389391199517782,
+      "grad_norm": 0.15996038913726807,
+      "learning_rate": 4.808270306022308e-05,
+      "loss": 1.4403,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6449668474984931,
+      "grad_norm": 0.1647796779870987,
+      "learning_rate": 4.801479041615046e-05,
+      "loss": 1.4844,
+      "step": 1070
+    },
+    {
+      "epoch": 0.650994575045208,
+      "grad_norm": 0.17005598545074463,
+      "learning_rate": 4.7945745520683025e-05,
+      "loss": 1.4628,
+      "step": 1080
+    },
+    {
+      "epoch": 0.6570223025919228,
+      "grad_norm": 0.1692131757736206,
+      "learning_rate": 4.787557177060077e-05,
+      "loss": 1.4842,
+      "step": 1090
+    },
+    {
+      "epoch": 0.6630500301386377,
+      "grad_norm": 0.16042789816856384,
+      "learning_rate": 4.780427261821957e-05,
+      "loss": 1.4338,
+      "step": 1100
+    },
+    {
+      "epoch": 0.6690777576853526,
+      "grad_norm": 0.1652117818593979,
+      "learning_rate": 4.773185157122136e-05,
+      "loss": 1.4623,
+      "step": 1110
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.17494116723537445,
+      "learning_rate": 4.7658312192481587e-05,
+      "loss": 1.4542,
+      "step": 1120
+    },
+    {
+      "epoch": 0.6811332127787824,
+      "grad_norm": 0.15841905772686005,
+      "learning_rate": 4.758365809989389e-05,
+      "loss": 1.4461,
+      "step": 1130
+    },
+    {
+      "epoch": 0.6871609403254972,
+      "grad_norm": 0.16204172372817993,
+      "learning_rate": 4.750789296619216e-05,
+      "loss": 1.4587,
+      "step": 1140
+    },
+    {
+      "epoch": 0.6931886678722122,
+      "grad_norm": 0.16820771992206573,
+      "learning_rate": 4.743102051876981e-05,
+      "loss": 1.4434,
+      "step": 1150
+    },
+    {
+      "epoch": 0.6992163954189271,
+      "grad_norm": 0.16587768495082855,
+      "learning_rate": 4.735304453949642e-05,
+      "loss": 1.4295,
+      "step": 1160
+    },
+    {
+      "epoch": 0.705244122965642,
+      "grad_norm": 0.17122101783752441,
+      "learning_rate": 4.727396886453168e-05,
+      "loss": 1.4533,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7112718505123569,
+      "grad_norm": 0.16851390898227692,
+      "learning_rate": 4.719379738413667e-05,
+      "loss": 1.4473,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.17578575015068054,
+      "learning_rate": 4.7112534042482456e-05,
+      "loss": 1.4211,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7233273056057866,
+      "grad_norm": 0.17948541045188904,
+      "learning_rate": 4.703018283745607e-05,
+      "loss": 1.4639,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7233273056057866,
+      "eval_loss": 1.4173197746276855,
+      "eval_runtime": 110.1107,
+      "eval_samples_per_second": 9.745,
+      "eval_steps_per_second": 2.443,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7293550331525015,
+      "grad_norm": 0.17049939930438995,
+      "learning_rate": 4.6946747820463784e-05,
+      "loss": 1.4517,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7353827606992164,
+      "grad_norm": 0.18009638786315918,
+      "learning_rate": 4.686223309623186e-05,
+      "loss": 1.4233,
+      "step": 1220
+    },
+    {
+      "epoch": 0.7414104882459313,
+      "grad_norm": 0.18773365020751953,
+      "learning_rate": 4.677664282260457e-05,
+      "loss": 1.4673,
+      "step": 1230
+    },
+    {
+      "epoch": 0.7474382157926461,
+      "grad_norm": 0.19085349142551422,
+      "learning_rate": 4.6689981210339636e-05,
+      "loss": 1.4211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.753465943339361,
+      "grad_norm": 0.17285512387752533,
+      "learning_rate": 4.660225252290111e-05,
+      "loss": 1.4312,
+      "step": 1250
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.18272553384304047,
+      "learning_rate": 4.651346107624959e-05,
+      "loss": 1.448,
+      "step": 1260
+    },
+    {
+      "epoch": 0.7655213984327909,
+      "grad_norm": 0.1952652931213379,
+      "learning_rate": 4.642361123862991e-05,
+      "loss": 1.4437,
+      "step": 1270
+    },
+    {
+      "epoch": 0.7715491259795058,
+      "grad_norm": 0.18138745427131653,
+      "learning_rate": 4.6332707430356214e-05,
+      "loss": 1.4338,
+      "step": 1280
+    },
+    {
+      "epoch": 0.7775768535262206,
+      "grad_norm": 0.18755076825618744,
+      "learning_rate": 4.6240754123594555e-05,
+      "loss": 1.4129,
+      "step": 1290
+    },
+    {
+      "epoch": 0.7836045810729355,
+      "grad_norm": 0.18552865087985992,
+      "learning_rate": 4.6147755842142785e-05,
+      "loss": 1.4414,
+      "step": 1300
+    },
+    {
+      "epoch": 0.7896323086196504,
+      "grad_norm": 0.1893608421087265,
+      "learning_rate": 4.605371716120805e-05,
+      "loss": 1.4264,
+      "step": 1310
+    },
+    {
+      "epoch": 0.7956600361663653,
+      "grad_norm": 0.1964706927537918,
+      "learning_rate": 4.595864270718175e-05,
+      "loss": 1.4267,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.1870163083076477,
+      "learning_rate": 4.586253715741182e-05,
+      "loss": 1.4051,
+      "step": 1330
+    },
+    {
+      "epoch": 0.807715491259795,
+      "grad_norm": 0.19142790138721466,
+      "learning_rate": 4.5765405239972736e-05,
+      "loss": 1.3907,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8137432188065099,
+      "grad_norm": 0.20232027769088745,
+      "learning_rate": 4.566725173343285e-05,
+      "loss": 1.4212,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8197709463532248,
+      "grad_norm": 0.2128591388463974,
+      "learning_rate": 4.5568081466619286e-05,
+      "loss": 1.4435,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8257986738999398,
+      "grad_norm": 0.19985191524028778,
+      "learning_rate": 4.546789931838044e-05,
+      "loss": 1.41,
+      "step": 1370
+    },
+    {
+      "epoch": 0.8318264014466547,
+      "grad_norm": 0.20753121376037598,
+      "learning_rate": 4.53667102173459e-05,
+      "loss": 1.4132,
+      "step": 1380
+    },
+    {
+      "epoch": 0.8378541289933695,
+      "grad_norm": 0.20165076851844788,
+      "learning_rate": 4.5264519141684e-05,
+      "loss": 1.4049,
+      "step": 1390
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.20434871315956116,
+      "learning_rate": 4.5161331118856885e-05,
+      "loss": 1.4151,
+      "step": 1400
+    },
+    {
+      "epoch": 0.8499095840867993,
+      "grad_norm": 0.21217265725135803,
+      "learning_rate": 4.5057151225373226e-05,
+      "loss": 1.4253,
+      "step": 1410
+    },
+    {
+      "epoch": 0.8559373116335142,
+      "grad_norm": 0.20287655293941498,
+      "learning_rate": 4.495198458653843e-05,
+      "loss": 1.4054,
+      "step": 1420
+    },
+    {
+      "epoch": 0.861965039180229,
+      "grad_norm": 0.21480686962604523,
+      "learning_rate": 4.4845836376202496e-05,
+      "loss": 1.4149,
+      "step": 1430
+    },
+    {
+      "epoch": 0.8679927667269439,
+      "grad_norm": 0.21573026478290558,
+      "learning_rate": 4.473871181650549e-05,
+      "loss": 1.4029,
+      "step": 1440
+    },
+    {
+      "epoch": 0.8740204942736588,
+      "grad_norm": 0.20224231481552124,
+      "learning_rate": 4.463061617762064e-05,
+      "loss": 1.4278,
+      "step": 1450
+    },
+    {
+      "epoch": 0.8800482218203737,
+      "grad_norm": 0.21252532303333282,
+      "learning_rate": 4.452155477749503e-05,
+      "loss": 1.3834,
+      "step": 1460
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.2173161506652832,
+      "learning_rate": 4.441153298158803e-05,
+      "loss": 1.4152,
+      "step": 1470
+    },
+    {
+      "epoch": 0.8921036769138035,
+      "grad_norm": 0.2368983030319214,
+      "learning_rate": 4.430055620260726e-05,
+      "loss": 1.404,
+      "step": 1480
+    },
+    {
+      "epoch": 0.8981314044605184,
+      "grad_norm": 0.22046008706092834,
+      "learning_rate": 4.4188629900242376e-05,
+      "loss": 1.4049,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9041591320072333,
+      "grad_norm": 0.21961750090122223,
+      "learning_rate": 4.40757595808964e-05,
+      "loss": 1.3673,
+      "step": 1500
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 4977,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.72182230024192e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:276772c4ef2f26970b87a5557f24d7c42bdfe56a8d7f3d49ad7c2ec614711f92
+size 7032

checkpoint-1500/zero_to_fp32.py

@@ -0,0 +1,674 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import json
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering Sharded Weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        state_dict_split = split_torch_state_dict_into_shards(state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors}
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard, output_path)
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

checkpoint-2000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen2.5-7B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.12.0

checkpoint-2000/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "q_proj",
+    "up_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": true
+}

checkpoint-2000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d51fdca17c4377c80b28b10d2b1716ae6dc1835f2e4c04ad8566e9e646dd2c96
+size 645976488

checkpoint-2000/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-2000/latest

@@ -0,0 +1 @@
+global_step2000

checkpoint-2000/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-2000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-2000/trainer_state.json

@@ -0,0 +1,1473 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.2055455093429777,
+  "eval_steps": 400,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006027727546714889,
+      "grad_norm": 6.095676898956299,
+      "learning_rate": 1.0040160642570282e-06,
+      "loss": 1.8585,
+      "step": 10
+    },
+    {
+      "epoch": 0.012055455093429777,
+      "grad_norm": 0.6981305480003357,
+      "learning_rate": 2.0080321285140564e-06,
+      "loss": 1.8607,
+      "step": 20
+    },
+    {
+      "epoch": 0.018083182640144666,
+      "grad_norm": 0.33439213037490845,
+      "learning_rate": 3.0120481927710846e-06,
+      "loss": 1.8035,
+      "step": 30
+    },
+    {
+      "epoch": 0.024110910186859555,
+      "grad_norm": 0.2445673942565918,
+      "learning_rate": 4.016064257028113e-06,
+      "loss": 1.7827,
+      "step": 40
+    },
+    {
+      "epoch": 0.030138637733574444,
+      "grad_norm": 0.22277717292308807,
+      "learning_rate": 5.020080321285141e-06,
+      "loss": 1.7306,
+      "step": 50
+    },
+    {
+      "epoch": 0.03616636528028933,
+      "grad_norm": 0.16803288459777832,
+      "learning_rate": 6.024096385542169e-06,
+      "loss": 1.728,
+      "step": 60
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.1401725709438324,
+      "learning_rate": 7.028112449799197e-06,
+      "loss": 1.6623,
+      "step": 70
+    },
+    {
+      "epoch": 0.04822182037371911,
+      "grad_norm": 0.14054332673549652,
+      "learning_rate": 8.032128514056226e-06,
+      "loss": 1.6226,
+      "step": 80
+    },
+    {
+      "epoch": 0.054249547920433995,
+      "grad_norm": 0.1356767863035202,
+      "learning_rate": 9.036144578313253e-06,
+      "loss": 1.6497,
+      "step": 90
+    },
+    {
+      "epoch": 0.06027727546714889,
+      "grad_norm": 0.14919587969779968,
+      "learning_rate": 1.0040160642570281e-05,
+      "loss": 1.6452,
+      "step": 100
+    },
+    {
+      "epoch": 0.06630500301386377,
+      "grad_norm": 0.16640327870845795,
+      "learning_rate": 1.104417670682731e-05,
+      "loss": 1.6311,
+      "step": 110
+    },
+    {
+      "epoch": 0.07233273056057866,
+      "grad_norm": 0.19770000874996185,
+      "learning_rate": 1.2048192771084338e-05,
+      "loss": 1.5925,
+      "step": 120
+    },
+    {
+      "epoch": 0.07836045810729356,
+      "grad_norm": 0.20399077236652374,
+      "learning_rate": 1.3052208835341367e-05,
+      "loss": 1.5825,
+      "step": 130
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.1512237936258316,
+      "learning_rate": 1.4056224899598394e-05,
+      "loss": 1.5478,
+      "step": 140
+    },
+    {
+      "epoch": 0.09041591320072333,
+      "grad_norm": 0.14794760942459106,
+      "learning_rate": 1.5060240963855424e-05,
+      "loss": 1.5252,
+      "step": 150
+    },
+    {
+      "epoch": 0.09644364074743822,
+      "grad_norm": 0.14903859794139862,
+      "learning_rate": 1.606425702811245e-05,
+      "loss": 1.5427,
+      "step": 160
+    },
+    {
+      "epoch": 0.1024713682941531,
+      "grad_norm": 0.1387884020805359,
+      "learning_rate": 1.706827309236948e-05,
+      "loss": 1.554,
+      "step": 170
+    },
+    {
+      "epoch": 0.10849909584086799,
+      "grad_norm": 0.14016373455524445,
+      "learning_rate": 1.8072289156626505e-05,
+      "loss": 1.5415,
+      "step": 180
+    },
+    {
+      "epoch": 0.11452682338758288,
+      "grad_norm": 0.14568917453289032,
+      "learning_rate": 1.9076305220883535e-05,
+      "loss": 1.5516,
+      "step": 190
+    },
+    {
+      "epoch": 0.12055455093429777,
+      "grad_norm": 0.14068633317947388,
+      "learning_rate": 2.0080321285140562e-05,
+      "loss": 1.5437,
+      "step": 200
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.1428183913230896,
+      "learning_rate": 2.1084337349397593e-05,
+      "loss": 1.5315,
+      "step": 210
+    },
+    {
+      "epoch": 0.13261000602772754,
+      "grad_norm": 0.1416633427143097,
+      "learning_rate": 2.208835341365462e-05,
+      "loss": 1.5276,
+      "step": 220
+    },
+    {
+      "epoch": 0.13863773357444242,
+      "grad_norm": 0.14373181760311127,
+      "learning_rate": 2.309236947791165e-05,
+      "loss": 1.5275,
+      "step": 230
+    },
+    {
+      "epoch": 0.14466546112115733,
+      "grad_norm": 0.16374213993549347,
+      "learning_rate": 2.4096385542168677e-05,
+      "loss": 1.5528,
+      "step": 240
+    },
+    {
+      "epoch": 0.1506931886678722,
+      "grad_norm": 0.13819484412670135,
+      "learning_rate": 2.5100401606425704e-05,
+      "loss": 1.5389,
+      "step": 250
+    },
+    {
+      "epoch": 0.1567209162145871,
+      "grad_norm": 0.14697293937206268,
+      "learning_rate": 2.6104417670682734e-05,
+      "loss": 1.5229,
+      "step": 260
+    },
+    {
+      "epoch": 0.162748643761302,
+      "grad_norm": 0.14389023184776306,
+      "learning_rate": 2.7108433734939758e-05,
+      "loss": 1.5376,
+      "step": 270
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.14889676868915558,
+      "learning_rate": 2.8112449799196788e-05,
+      "loss": 1.5508,
+      "step": 280
+    },
+    {
+      "epoch": 0.17480409885473178,
+      "grad_norm": 0.15245158970355988,
+      "learning_rate": 2.911646586345382e-05,
+      "loss": 1.5342,
+      "step": 290
+    },
+    {
+      "epoch": 0.18083182640144665,
+      "grad_norm": 0.143811896443367,
+      "learning_rate": 3.012048192771085e-05,
+      "loss": 1.5123,
+      "step": 300
+    },
+    {
+      "epoch": 0.18685955394816153,
+      "grad_norm": 0.14010299742221832,
+      "learning_rate": 3.112449799196787e-05,
+      "loss": 1.5355,
+      "step": 310
+    },
+    {
+      "epoch": 0.19288728149487644,
+      "grad_norm": 0.14191724359989166,
+      "learning_rate": 3.21285140562249e-05,
+      "loss": 1.5052,
+      "step": 320
+    },
+    {
+      "epoch": 0.19891500904159132,
+      "grad_norm": 0.13444629311561584,
+      "learning_rate": 3.313253012048193e-05,
+      "loss": 1.5357,
+      "step": 330
+    },
+    {
+      "epoch": 0.2049427365883062,
+      "grad_norm": 0.15248367190361023,
+      "learning_rate": 3.413654618473896e-05,
+      "loss": 1.5171,
+      "step": 340
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.16744668781757355,
+      "learning_rate": 3.5140562248995983e-05,
+      "loss": 1.5379,
+      "step": 350
+    },
+    {
+      "epoch": 0.21699819168173598,
+      "grad_norm": 0.1389068365097046,
+      "learning_rate": 3.614457831325301e-05,
+      "loss": 1.5142,
+      "step": 360
+    },
+    {
+      "epoch": 0.2230259192284509,
+      "grad_norm": 0.13487495481967926,
+      "learning_rate": 3.7148594377510044e-05,
+      "loss": 1.5189,
+      "step": 370
+    },
+    {
+      "epoch": 0.22905364677516576,
+      "grad_norm": 0.13311463594436646,
+      "learning_rate": 3.815261044176707e-05,
+      "loss": 1.5166,
+      "step": 380
+    },
+    {
+      "epoch": 0.23508137432188064,
+      "grad_norm": 0.1321990191936493,
+      "learning_rate": 3.91566265060241e-05,
+      "loss": 1.5218,
+      "step": 390
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "grad_norm": 0.1350693255662918,
+      "learning_rate": 4.0160642570281125e-05,
+      "loss": 1.54,
+      "step": 400
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "eval_loss": 1.4954419136047363,
+      "eval_runtime": 110.0553,
+      "eval_samples_per_second": 9.75,
+      "eval_steps_per_second": 2.444,
+      "step": 400
+    },
+    {
+      "epoch": 0.24713682941531043,
+      "grad_norm": 0.1386946737766266,
+      "learning_rate": 4.116465863453816e-05,
+      "loss": 1.5307,
+      "step": 410
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.1321861445903778,
+      "learning_rate": 4.2168674698795186e-05,
+      "loss": 1.5036,
+      "step": 420
+    },
+    {
+      "epoch": 0.2591922845087402,
+      "grad_norm": 0.13090862333774567,
+      "learning_rate": 4.317269076305221e-05,
+      "loss": 1.5036,
+      "step": 430
+    },
+    {
+      "epoch": 0.2652200120554551,
+      "grad_norm": 0.14390167593955994,
+      "learning_rate": 4.417670682730924e-05,
+      "loss": 1.5245,
+      "step": 440
+    },
+    {
+      "epoch": 0.27124773960216997,
+      "grad_norm": 0.14252203702926636,
+      "learning_rate": 4.5180722891566266e-05,
+      "loss": 1.5361,
+      "step": 450
+    },
+    {
+      "epoch": 0.27727546714888485,
+      "grad_norm": 0.1380358338356018,
+      "learning_rate": 4.61847389558233e-05,
+      "loss": 1.5379,
+      "step": 460
+    },
+    {
+      "epoch": 0.2833031946955998,
+      "grad_norm": 0.17792685329914093,
+      "learning_rate": 4.718875502008032e-05,
+      "loss": 1.5107,
+      "step": 470
+    },
+    {
+      "epoch": 0.28933092224231466,
+      "grad_norm": 0.1495760679244995,
+      "learning_rate": 4.8192771084337354e-05,
+      "loss": 1.5063,
+      "step": 480
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.12926514446735382,
+      "learning_rate": 4.919678714859438e-05,
+      "loss": 1.5149,
+      "step": 490
+    },
+    {
+      "epoch": 0.3013863773357444,
+      "grad_norm": 0.12767350673675537,
+      "learning_rate": 4.999997540156068e-05,
+      "loss": 1.5052,
+      "step": 500
+    },
+    {
+      "epoch": 0.3074141048824593,
+      "grad_norm": 0.12568402290344238,
+      "learning_rate": 4.999911446126714e-05,
+      "loss": 1.541,
+      "step": 510
+    },
+    {
+      "epoch": 0.3134418324291742,
+      "grad_norm": 0.1353764533996582,
+      "learning_rate": 4.9997023647413776e-05,
+      "loss": 1.5218,
+      "step": 520
+    },
+    {
+      "epoch": 0.3194695599758891,
+      "grad_norm": 2.775493621826172,
+      "learning_rate": 4.999370306286171e-05,
+      "loss": 1.5301,
+      "step": 530
+    },
+    {
+      "epoch": 0.325497287522604,
+      "grad_norm": 0.12400302290916443,
+      "learning_rate": 4.998915287097269e-05,
+      "loss": 1.4685,
+      "step": 540
+    },
+    {
+      "epoch": 0.33152501506931886,
+      "grad_norm": 0.12950482964515686,
+      "learning_rate": 4.998337329560107e-05,
+      "loss": 1.5264,
+      "step": 550
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.12773030996322632,
+      "learning_rate": 4.99763646210828e-05,
+      "loss": 1.4937,
+      "step": 560
+    },
+    {
+      "epoch": 0.3435804701627486,
+      "grad_norm": 0.12648645043373108,
+      "learning_rate": 4.9968127192221436e-05,
+      "loss": 1.5126,
+      "step": 570
+    },
+    {
+      "epoch": 0.34960819770946355,
+      "grad_norm": 0.12337373942136765,
+      "learning_rate": 4.995866141427117e-05,
+      "loss": 1.4816,
+      "step": 580
+    },
+    {
+      "epoch": 0.35563592525617843,
+      "grad_norm": 0.1327151358127594,
+      "learning_rate": 4.99479677529169e-05,
+      "loss": 1.5131,
+      "step": 590
+    },
+    {
+      "epoch": 0.3616636528028933,
+      "grad_norm": 0.13992607593536377,
+      "learning_rate": 4.9936046734251313e-05,
+      "loss": 1.5089,
+      "step": 600
+    },
+    {
+      "epoch": 0.3676913803496082,
+      "grad_norm": 0.12979301810264587,
+      "learning_rate": 4.992289894474901e-05,
+      "loss": 1.5032,
+      "step": 610
+    },
+    {
+      "epoch": 0.37371910789632307,
+      "grad_norm": 0.1280072033405304,
+      "learning_rate": 4.990852503123765e-05,
+      "loss": 1.5012,
+      "step": 620
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.1306256204843521,
+      "learning_rate": 4.989292570086612e-05,
+      "loss": 1.4564,
+      "step": 630
+    },
+    {
+      "epoch": 0.3857745629897529,
+      "grad_norm": 0.13198673725128174,
+      "learning_rate": 4.987610172106978e-05,
+      "loss": 1.5106,
+      "step": 640
+    },
+    {
+      "epoch": 0.39180229053646776,
+      "grad_norm": 0.1281399428844452,
+      "learning_rate": 4.9858053919532646e-05,
+      "loss": 1.4834,
+      "step": 650
+    },
+    {
+      "epoch": 0.39783001808318263,
+      "grad_norm": 0.13263563811779022,
+      "learning_rate": 4.983878318414674e-05,
+      "loss": 1.5208,
+      "step": 660
+    },
+    {
+      "epoch": 0.4038577456298975,
+      "grad_norm": 0.13027793169021606,
+      "learning_rate": 4.981829046296836e-05,
+      "loss": 1.4892,
+      "step": 670
+    },
+    {
+      "epoch": 0.4098854731766124,
+      "grad_norm": 0.13578078150749207,
+      "learning_rate": 4.9796576764171446e-05,
+      "loss": 1.4874,
+      "step": 680
+    },
+    {
+      "epoch": 0.4159132007233273,
+      "grad_norm": 0.13204683363437653,
+      "learning_rate": 4.9773643155998e-05,
+      "loss": 1.4867,
+      "step": 690
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.13477781414985657,
+      "learning_rate": 4.974949076670553e-05,
+      "loss": 1.5159,
+      "step": 700
+    },
+    {
+      "epoch": 0.4279686558167571,
+      "grad_norm": 0.13009533286094666,
+      "learning_rate": 4.972412078451151e-05,
+      "loss": 1.4806,
+      "step": 710
+    },
+    {
+      "epoch": 0.43399638336347196,
+      "grad_norm": 0.1321401745080948,
+      "learning_rate": 4.9697534457534966e-05,
+      "loss": 1.4641,
+      "step": 720
+    },
+    {
+      "epoch": 0.44002411091018684,
+      "grad_norm": 0.13463927805423737,
+      "learning_rate": 4.966973309373506e-05,
+      "loss": 1.4924,
+      "step": 730
+    },
+    {
+      "epoch": 0.4460518384569018,
+      "grad_norm": 0.1337679624557495,
+      "learning_rate": 4.9640718060846723e-05,
+      "loss": 1.4805,
+      "step": 740
+    },
+    {
+      "epoch": 0.45207956600361665,
+      "grad_norm": 0.1317020207643509,
+      "learning_rate": 4.96104907863134e-05,
+      "loss": 1.4678,
+      "step": 750
+    },
+    {
+      "epoch": 0.45810729355033153,
+      "grad_norm": 0.13243788480758667,
+      "learning_rate": 4.957905275721678e-05,
+      "loss": 1.4794,
+      "step": 760
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.13113069534301758,
+      "learning_rate": 4.954640552020369e-05,
+      "loss": 1.4734,
+      "step": 770
+    },
+    {
+      "epoch": 0.4701627486437613,
+      "grad_norm": 0.13361573219299316,
+      "learning_rate": 4.9512550681409946e-05,
+      "loss": 1.4803,
+      "step": 780
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.13844482600688934,
+      "learning_rate": 4.9477489906381414e-05,
+      "loss": 1.4665,
+      "step": 790
+    },
+    {
+      "epoch": 0.4822182037371911,
+      "grad_norm": 0.13865956664085388,
+      "learning_rate": 4.944122491999198e-05,
+      "loss": 1.4872,
+      "step": 800
+    },
+    {
+      "epoch": 0.4822182037371911,
+      "eval_loss": 1.4607902765274048,
+      "eval_runtime": 110.1311,
+      "eval_samples_per_second": 9.743,
+      "eval_steps_per_second": 2.443,
+      "step": 800
+    },
+    {
+      "epoch": 0.488245931283906,
+      "grad_norm": 0.13776348531246185,
+      "learning_rate": 4.9403757506358775e-05,
+      "loss": 1.5016,
+      "step": 810
+    },
+    {
+      "epoch": 0.49427365883062085,
+      "grad_norm": 0.14617718756198883,
+      "learning_rate": 4.936508950875432e-05,
+      "loss": 1.4857,
+      "step": 820
+    },
+    {
+      "epoch": 0.5003013863773358,
+      "grad_norm": 0.14705806970596313,
+      "learning_rate": 4.932522282951594e-05,
+      "loss": 1.4909,
+      "step": 830
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.14169825613498688,
+      "learning_rate": 4.928415942995208e-05,
+      "loss": 1.5106,
+      "step": 840
+    },
+    {
+      "epoch": 0.5123568414707655,
+      "grad_norm": 0.14307105541229248,
+      "learning_rate": 4.924190133024589e-05,
+      "loss": 1.492,
+      "step": 850
+    },
+    {
+      "epoch": 0.5183845690174804,
+      "grad_norm": 0.1347261369228363,
+      "learning_rate": 4.919845060935577e-05,
+      "loss": 1.5063,
+      "step": 860
+    },
+    {
+      "epoch": 0.5244122965641953,
+      "grad_norm": 0.1423402726650238,
+      "learning_rate": 4.915380940491315e-05,
+      "loss": 1.4682,
+      "step": 870
+    },
+    {
+      "epoch": 0.5304400241109102,
+      "grad_norm": 0.14149978756904602,
+      "learning_rate": 4.910797991311733e-05,
+      "loss": 1.4436,
+      "step": 880
+    },
+    {
+      "epoch": 0.5364677516576251,
+      "grad_norm": 0.1491919308900833,
+      "learning_rate": 4.906096438862735e-05,
+      "loss": 1.4851,
+      "step": 890
+    },
+    {
+      "epoch": 0.5424954792043399,
+      "grad_norm": 0.14350004494190216,
+      "learning_rate": 4.9012765144451186e-05,
+      "loss": 1.4546,
+      "step": 900
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.14726224541664124,
+      "learning_rate": 4.896338455183185e-05,
+      "loss": 1.473,
+      "step": 910
+    },
+    {
+      "epoch": 0.5545509342977697,
+      "grad_norm": 0.1460171639919281,
+      "learning_rate": 4.8912825040130814e-05,
+      "loss": 1.4802,
+      "step": 920
+    },
+    {
+      "epoch": 0.5605786618444847,
+      "grad_norm": 0.15167438983917236,
+      "learning_rate": 4.886108909670845e-05,
+      "loss": 1.4542,
+      "step": 930
+    },
+    {
+      "epoch": 0.5666063893911996,
+      "grad_norm": 0.15603500604629517,
+      "learning_rate": 4.880817926680167e-05,
+      "loss": 1.4565,
+      "step": 940
+    },
+    {
+      "epoch": 0.5726341169379144,
+      "grad_norm": 0.14839062094688416,
+      "learning_rate": 4.8754098153398706e-05,
+      "loss": 1.4833,
+      "step": 950
+    },
+    {
+      "epoch": 0.5786618444846293,
+      "grad_norm": 0.14976999163627625,
+      "learning_rate": 4.869884841711107e-05,
+      "loss": 1.4957,
+      "step": 960
+    },
+    {
+      "epoch": 0.5846895720313442,
+      "grad_norm": 0.15414153039455414,
+      "learning_rate": 4.8642432776042616e-05,
+      "loss": 1.4816,
+      "step": 970
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.15507987141609192,
+      "learning_rate": 4.858485400565589e-05,
+      "loss": 1.4728,
+      "step": 980
+    },
+    {
+      "epoch": 0.596745027124774,
+      "grad_norm": 0.1561218947172165,
+      "learning_rate": 4.85261149386355e-05,
+      "loss": 1.4936,
+      "step": 990
+    },
+    {
+      "epoch": 0.6027727546714888,
+      "grad_norm": 0.15994535386562347,
+      "learning_rate": 4.846621846474885e-05,
+      "loss": 1.4705,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6088004822182037,
+      "grad_norm": 0.14968565106391907,
+      "learning_rate": 4.840516753070387e-05,
+      "loss": 1.4638,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6148282097649186,
+      "grad_norm": 0.1536279022693634,
+      "learning_rate": 4.834296514000416e-05,
+      "loss": 1.4757,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6208559373116335,
+      "grad_norm": 0.1513814926147461,
+      "learning_rate": 4.827961435280112e-05,
+      "loss": 1.4548,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6268836648583485,
+      "grad_norm": 0.16222743690013885,
+      "learning_rate": 4.821511828574349e-05,
+      "loss": 1.4643,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.1569400429725647,
+      "learning_rate": 4.814948011182395e-05,
+      "loss": 1.4459,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6389391199517782,
+      "grad_norm": 0.15996038913726807,
+      "learning_rate": 4.808270306022308e-05,
+      "loss": 1.4403,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6449668474984931,
+      "grad_norm": 0.1647796779870987,
+      "learning_rate": 4.801479041615046e-05,
+      "loss": 1.4844,
+      "step": 1070
+    },
+    {
+      "epoch": 0.650994575045208,
+      "grad_norm": 0.17005598545074463,
+      "learning_rate": 4.7945745520683025e-05,
+      "loss": 1.4628,
+      "step": 1080
+    },
+    {
+      "epoch": 0.6570223025919228,
+      "grad_norm": 0.1692131757736206,
+      "learning_rate": 4.787557177060077e-05,
+      "loss": 1.4842,
+      "step": 1090
+    },
+    {
+      "epoch": 0.6630500301386377,
+      "grad_norm": 0.16042789816856384,
+      "learning_rate": 4.780427261821957e-05,
+      "loss": 1.4338,
+      "step": 1100
+    },
+    {
+      "epoch": 0.6690777576853526,
+      "grad_norm": 0.1652117818593979,
+      "learning_rate": 4.773185157122136e-05,
+      "loss": 1.4623,
+      "step": 1110
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.17494116723537445,
+      "learning_rate": 4.7658312192481587e-05,
+      "loss": 1.4542,
+      "step": 1120
+    },
+    {
+      "epoch": 0.6811332127787824,
+      "grad_norm": 0.15841905772686005,
+      "learning_rate": 4.758365809989389e-05,
+      "loss": 1.4461,
+      "step": 1130
+    },
+    {
+      "epoch": 0.6871609403254972,
+      "grad_norm": 0.16204172372817993,
+      "learning_rate": 4.750789296619216e-05,
+      "loss": 1.4587,
+      "step": 1140
+    },
+    {
+      "epoch": 0.6931886678722122,
+      "grad_norm": 0.16820771992206573,
+      "learning_rate": 4.743102051876981e-05,
+      "loss": 1.4434,
+      "step": 1150
+    },
+    {
+      "epoch": 0.6992163954189271,
+      "grad_norm": 0.16587768495082855,
+      "learning_rate": 4.735304453949642e-05,
+      "loss": 1.4295,
+      "step": 1160
+    },
+    {
+      "epoch": 0.705244122965642,
+      "grad_norm": 0.17122101783752441,
+      "learning_rate": 4.727396886453168e-05,
+      "loss": 1.4533,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7112718505123569,
+      "grad_norm": 0.16851390898227692,
+      "learning_rate": 4.719379738413667e-05,
+      "loss": 1.4473,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.17578575015068054,
+      "learning_rate": 4.7112534042482456e-05,
+      "loss": 1.4211,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7233273056057866,
+      "grad_norm": 0.17948541045188904,
+      "learning_rate": 4.703018283745607e-05,
+      "loss": 1.4639,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7233273056057866,
+      "eval_loss": 1.4173197746276855,
+      "eval_runtime": 110.1107,
+      "eval_samples_per_second": 9.745,
+      "eval_steps_per_second": 2.443,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7293550331525015,
+      "grad_norm": 0.17049939930438995,
+      "learning_rate": 4.6946747820463784e-05,
+      "loss": 1.4517,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7353827606992164,
+      "grad_norm": 0.18009638786315918,
+      "learning_rate": 4.686223309623186e-05,
+      "loss": 1.4233,
+      "step": 1220
+    },
+    {
+      "epoch": 0.7414104882459313,
+      "grad_norm": 0.18773365020751953,
+      "learning_rate": 4.677664282260457e-05,
+      "loss": 1.4673,
+      "step": 1230
+    },
+    {
+      "epoch": 0.7474382157926461,
+      "grad_norm": 0.19085349142551422,
+      "learning_rate": 4.6689981210339636e-05,
+      "loss": 1.4211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.753465943339361,
+      "grad_norm": 0.17285512387752533,
+      "learning_rate": 4.660225252290111e-05,
+      "loss": 1.4312,
+      "step": 1250
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.18272553384304047,
+      "learning_rate": 4.651346107624959e-05,
+      "loss": 1.448,
+      "step": 1260
+    },
+    {
+      "epoch": 0.7655213984327909,
+      "grad_norm": 0.1952652931213379,
+      "learning_rate": 4.642361123862991e-05,
+      "loss": 1.4437,
+      "step": 1270
+    },
+    {
+      "epoch": 0.7715491259795058,
+      "grad_norm": 0.18138745427131653,
+      "learning_rate": 4.6332707430356214e-05,
+      "loss": 1.4338,
+      "step": 1280
+    },
+    {
+      "epoch": 0.7775768535262206,
+      "grad_norm": 0.18755076825618744,
+      "learning_rate": 4.6240754123594555e-05,
+      "loss": 1.4129,
+      "step": 1290
+    },
+    {
+      "epoch": 0.7836045810729355,
+      "grad_norm": 0.18552865087985992,
+      "learning_rate": 4.6147755842142785e-05,
+      "loss": 1.4414,
+      "step": 1300
+    },
+    {
+      "epoch": 0.7896323086196504,
+      "grad_norm": 0.1893608421087265,
+      "learning_rate": 4.605371716120805e-05,
+      "loss": 1.4264,
+      "step": 1310
+    },
+    {
+      "epoch": 0.7956600361663653,
+      "grad_norm": 0.1964706927537918,
+      "learning_rate": 4.595864270718175e-05,
+      "loss": 1.4267,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.1870163083076477,
+      "learning_rate": 4.586253715741182e-05,
+      "loss": 1.4051,
+      "step": 1330
+    },
+    {
+      "epoch": 0.807715491259795,
+      "grad_norm": 0.19142790138721466,
+      "learning_rate": 4.5765405239972736e-05,
+      "loss": 1.3907,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8137432188065099,
+      "grad_norm": 0.20232027769088745,
+      "learning_rate": 4.566725173343285e-05,
+      "loss": 1.4212,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8197709463532248,
+      "grad_norm": 0.2128591388463974,
+      "learning_rate": 4.5568081466619286e-05,
+      "loss": 1.4435,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8257986738999398,
+      "grad_norm": 0.19985191524028778,
+      "learning_rate": 4.546789931838044e-05,
+      "loss": 1.41,
+      "step": 1370
+    },
+    {
+      "epoch": 0.8318264014466547,
+      "grad_norm": 0.20753121376037598,
+      "learning_rate": 4.53667102173459e-05,
+      "loss": 1.4132,
+      "step": 1380
+    },
+    {
+      "epoch": 0.8378541289933695,
+      "grad_norm": 0.20165076851844788,
+      "learning_rate": 4.5264519141684e-05,
+      "loss": 1.4049,
+      "step": 1390
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.20434871315956116,
+      "learning_rate": 4.5161331118856885e-05,
+      "loss": 1.4151,
+      "step": 1400
+    },
+    {
+      "epoch": 0.8499095840867993,
+      "grad_norm": 0.21217265725135803,
+      "learning_rate": 4.5057151225373226e-05,
+      "loss": 1.4253,
+      "step": 1410
+    },
+    {
+      "epoch": 0.8559373116335142,
+      "grad_norm": 0.20287655293941498,
+      "learning_rate": 4.495198458653843e-05,
+      "loss": 1.4054,
+      "step": 1420
+    },
+    {
+      "epoch": 0.861965039180229,
+      "grad_norm": 0.21480686962604523,
+      "learning_rate": 4.4845836376202496e-05,
+      "loss": 1.4149,
+      "step": 1430
+    },
+    {
+      "epoch": 0.8679927667269439,
+      "grad_norm": 0.21573026478290558,
+      "learning_rate": 4.473871181650549e-05,
+      "loss": 1.4029,
+      "step": 1440
+    },
+    {
+      "epoch": 0.8740204942736588,
+      "grad_norm": 0.20224231481552124,
+      "learning_rate": 4.463061617762064e-05,
+      "loss": 1.4278,
+      "step": 1450
+    },
+    {
+      "epoch": 0.8800482218203737,
+      "grad_norm": 0.21252532303333282,
+      "learning_rate": 4.452155477749503e-05,
+      "loss": 1.3834,
+      "step": 1460
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.2173161506652832,
+      "learning_rate": 4.441153298158803e-05,
+      "loss": 1.4152,
+      "step": 1470
+    },
+    {
+      "epoch": 0.8921036769138035,
+      "grad_norm": 0.2368983030319214,
+      "learning_rate": 4.430055620260726e-05,
+      "loss": 1.404,
+      "step": 1480
+    },
+    {
+      "epoch": 0.8981314044605184,
+      "grad_norm": 0.22046008706092834,
+      "learning_rate": 4.4188629900242376e-05,
+      "loss": 1.4049,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9041591320072333,
+      "grad_norm": 0.21961750090122223,
+      "learning_rate": 4.40757595808964e-05,
+      "loss": 1.3673,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9101868595539482,
+      "grad_norm": 0.21616864204406738,
+      "learning_rate": 4.396195079741489e-05,
+      "loss": 1.3827,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9162145871006631,
+      "grad_norm": 0.22866834700107574,
+      "learning_rate": 4.384720914881273e-05,
+      "loss": 1.3799,
+      "step": 1520
+    },
+    {
+      "epoch": 0.9222423146473779,
+      "grad_norm": 0.23651176691055298,
+      "learning_rate": 4.373154027999865e-05,
+      "loss": 1.3674,
+      "step": 1530
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.2270604968070984,
+      "learning_rate": 4.3614949881497555e-05,
+      "loss": 1.404,
+      "step": 1540
+    },
+    {
+      "epoch": 0.9342977697408077,
+      "grad_norm": 0.24052606523036957,
+      "learning_rate": 4.349744368917057e-05,
+      "loss": 1.3926,
+      "step": 1550
+    },
+    {
+      "epoch": 0.9403254972875226,
+      "grad_norm": 0.25563594698905945,
+      "learning_rate": 4.337902748393281e-05,
+      "loss": 1.3814,
+      "step": 1560
+    },
+    {
+      "epoch": 0.9463532248342374,
+      "grad_norm": 0.23579391837120056,
+      "learning_rate": 4.3259707091469036e-05,
+      "loss": 1.4086,
+      "step": 1570
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.24091574549674988,
+      "learning_rate": 4.313948838194701e-05,
+      "loss": 1.3929,
+      "step": 1580
+    },
+    {
+      "epoch": 0.9584086799276673,
+      "grad_norm": 0.24928399920463562,
+      "learning_rate": 4.301837726972871e-05,
+      "loss": 1.4287,
+      "step": 1590
+    },
+    {
+      "epoch": 0.9644364074743822,
+      "grad_norm": 0.2525610625743866,
+      "learning_rate": 4.289637971307938e-05,
+      "loss": 1.3837,
+      "step": 1600
+    },
+    {
+      "epoch": 0.9644364074743822,
+      "eval_loss": 1.3578946590423584,
+      "eval_runtime": 109.9596,
+      "eval_samples_per_second": 9.758,
+      "eval_steps_per_second": 2.446,
+      "step": 1600
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.23288561403751373,
+      "learning_rate": 4.27735017138744e-05,
+      "loss": 1.3653,
+      "step": 1610
+    },
+    {
+      "epoch": 0.976491862567812,
+      "grad_norm": 0.23934006690979004,
+      "learning_rate": 4.264974931730396e-05,
+      "loss": 1.3944,
+      "step": 1620
+    },
+    {
+      "epoch": 0.9825195901145268,
+      "grad_norm": 0.2440434992313385,
+      "learning_rate": 4.252512861157577e-05,
+      "loss": 1.3721,
+      "step": 1630
+    },
+    {
+      "epoch": 0.9885473176612417,
+      "grad_norm": 0.24858222901821136,
+      "learning_rate": 4.239964572761542e-05,
+      "loss": 1.3672,
+      "step": 1640
+    },
+    {
+      "epoch": 0.9945750452079566,
+      "grad_norm": 0.255058616399765,
+      "learning_rate": 4.2273306838764825e-05,
+      "loss": 1.3768,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0006027727546716,
+      "grad_norm": 0.2636558413505554,
+      "learning_rate": 4.2146118160478505e-05,
+      "loss": 1.3438,
+      "step": 1660
+    },
+    {
+      "epoch": 1.0066305003013865,
+      "grad_norm": 0.2660534977912903,
+      "learning_rate": 4.2018085950017804e-05,
+      "loss": 1.3228,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "grad_norm": 0.2813737094402313,
+      "learning_rate": 4.1889216506143056e-05,
+      "loss": 1.3205,
+      "step": 1680
+    },
+    {
+      "epoch": 1.0186859553948162,
+      "grad_norm": 0.2968331277370453,
+      "learning_rate": 4.1759516168803694e-05,
+      "loss": 1.3069,
+      "step": 1690
+    },
+    {
+      "epoch": 1.024713682941531,
+      "grad_norm": 0.28205934166908264,
+      "learning_rate": 4.1628991318826365e-05,
+      "loss": 1.3161,
+      "step": 1700
+    },
+    {
+      "epoch": 1.030741410488246,
+      "grad_norm": 0.27789610624313354,
+      "learning_rate": 4.1497648377601005e-05,
+      "loss": 1.3354,
+      "step": 1710
+    },
+    {
+      "epoch": 1.0367691380349608,
+      "grad_norm": 0.27981898188591003,
+      "learning_rate": 4.136549380676491e-05,
+      "loss": 1.3201,
+      "step": 1720
+    },
+    {
+      "epoch": 1.0427968655816757,
+      "grad_norm": 0.2824476957321167,
+      "learning_rate": 4.123253410788488e-05,
+      "loss": 1.3295,
+      "step": 1730
+    },
+    {
+      "epoch": 1.0488245931283906,
+      "grad_norm": 0.279563307762146,
+      "learning_rate": 4.1098775822137346e-05,
+      "loss": 1.2988,
+      "step": 1740
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "grad_norm": 0.30854421854019165,
+      "learning_rate": 4.096422552998654e-05,
+      "loss": 1.3295,
+      "step": 1750
+    },
+    {
+      "epoch": 1.0608800482218204,
+      "grad_norm": 0.292086124420166,
+      "learning_rate": 4.082888985086084e-05,
+      "loss": 1.2776,
+      "step": 1760
+    },
+    {
+      "epoch": 1.0669077757685352,
+      "grad_norm": 0.2996542155742645,
+      "learning_rate": 4.069277544282699e-05,
+      "loss": 1.3308,
+      "step": 1770
+    },
+    {
+      "epoch": 1.0729355033152501,
+      "grad_norm": 0.2958160936832428,
+      "learning_rate": 4.055588900226267e-05,
+      "loss": 1.3138,
+      "step": 1780
+    },
+    {
+      "epoch": 1.078963230861965,
+      "grad_norm": 0.29136922955513,
+      "learning_rate": 4.041823726352696e-05,
+      "loss": 1.3303,
+      "step": 1790
+    },
+    {
+      "epoch": 1.0849909584086799,
+      "grad_norm": 0.29927417635917664,
+      "learning_rate": 4.02798269986291e-05,
+      "loss": 1.2884,
+      "step": 1800
+    },
+    {
+      "epoch": 1.0910186859553948,
+      "grad_norm": 0.30471071600914,
+      "learning_rate": 4.014066501689531e-05,
+      "loss": 1.3305,
+      "step": 1810
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "grad_norm": 0.2950434684753418,
+      "learning_rate": 4.0000758164633745e-05,
+      "loss": 1.3009,
+      "step": 1820
+    },
+    {
+      "epoch": 1.1030741410488245,
+      "grad_norm": 0.30433064699172974,
+      "learning_rate": 3.986011332479777e-05,
+      "loss": 1.3154,
+      "step": 1830
+    },
+    {
+      "epoch": 1.1091018685955394,
+      "grad_norm": 0.32296738028526306,
+      "learning_rate": 3.971873741664726e-05,
+      "loss": 1.2916,
+      "step": 1840
+    },
+    {
+      "epoch": 1.1151295961422543,
+      "grad_norm": 0.30300599336624146,
+      "learning_rate": 3.957663739540823e-05,
+      "loss": 1.2712,
+      "step": 1850
+    },
+    {
+      "epoch": 1.1211573236889691,
+      "grad_norm": 0.3014506697654724,
+      "learning_rate": 3.9433820251930675e-05,
+      "loss": 1.2903,
+      "step": 1860
+    },
+    {
+      "epoch": 1.1271850512356842,
+      "grad_norm": 0.3053356409072876,
+      "learning_rate": 3.9290293012344624e-05,
+      "loss": 1.3064,
+      "step": 1870
+    },
+    {
+      "epoch": 1.1332127787823991,
+      "grad_norm": 0.31932970881462097,
+      "learning_rate": 3.914606273771446e-05,
+      "loss": 1.2805,
+      "step": 1880
+    },
+    {
+      "epoch": 1.139240506329114,
+      "grad_norm": 0.3033607006072998,
+      "learning_rate": 3.9001136523691584e-05,
+      "loss": 1.26,
+      "step": 1890
+    },
+    {
+      "epoch": 1.1452682338758289,
+      "grad_norm": 0.30965933203697205,
+      "learning_rate": 3.885552150016528e-05,
+      "loss": 1.2992,
+      "step": 1900
+    },
+    {
+      "epoch": 1.1512959614225438,
+      "grad_norm": 0.30874183773994446,
+      "learning_rate": 3.8709224830912024e-05,
+      "loss": 1.2718,
+      "step": 1910
+    },
+    {
+      "epoch": 1.1573236889692586,
+      "grad_norm": 0.31119504570961,
+      "learning_rate": 3.856225371324295e-05,
+      "loss": 1.3106,
+      "step": 1920
+    },
+    {
+      "epoch": 1.1633514165159735,
+      "grad_norm": 0.32276487350463867,
+      "learning_rate": 3.841461537764984e-05,
+      "loss": 1.2818,
+      "step": 1930
+    },
+    {
+      "epoch": 1.1693791440626884,
+      "grad_norm": 0.31319287419319153,
+      "learning_rate": 3.8266317087449385e-05,
+      "loss": 1.3154,
+      "step": 1940
+    },
+    {
+      "epoch": 1.1754068716094033,
+      "grad_norm": 0.3152843117713928,
+      "learning_rate": 3.811736613842588e-05,
+      "loss": 1.3068,
+      "step": 1950
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "grad_norm": 0.3310001492500305,
+      "learning_rate": 3.796776985847222e-05,
+      "loss": 1.2885,
+      "step": 1960
+    },
+    {
+      "epoch": 1.187462326702833,
+      "grad_norm": 0.335748553276062,
+      "learning_rate": 3.781753560722948e-05,
+      "loss": 1.282,
+      "step": 1970
+    },
+    {
+      "epoch": 1.193490054249548,
+      "grad_norm": 0.30873429775238037,
+      "learning_rate": 3.76666707757248e-05,
+      "loss": 1.288,
+      "step": 1980
+    },
+    {
+      "epoch": 1.1995177817962628,
+      "grad_norm": 0.336524099111557,
+      "learning_rate": 3.7515182786007774e-05,
+      "loss": 1.2613,
+      "step": 1990
+    },
+    {
+      "epoch": 1.2055455093429777,
+      "grad_norm": 0.337877094745636,
+      "learning_rate": 3.73630790907853e-05,
+      "loss": 1.2826,
+      "step": 2000
+    },
+    {
+      "epoch": 1.2055455093429777,
+      "eval_loss": 1.293273687362671,
+      "eval_runtime": 110.0977,
+      "eval_samples_per_second": 9.746,
+      "eval_steps_per_second": 2.443,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 4977,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.162909640032256e+19,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:276772c4ef2f26970b87a5557f24d7c42bdfe56a8d7f3d49ad7c2ec614711f92
+size 7032

checkpoint-2000/zero_to_fp32.py

@@ -0,0 +1,674 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import json
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering Sharded Weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        state_dict_split = split_torch_state_dict_into_shards(state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors}
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard, output_path)
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

checkpoint-500/README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen2.5-7B
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.12.0

checkpoint-500/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "q_proj",
+    "up_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": true
+}

checkpoint-500/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba9b182f3b0d351540a22af8bd990c9e2749139859a561b24dc1132229677db7
+size 645976488

checkpoint-500/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-500/latest

@@ -0,0 +1 @@
+global_step500

checkpoint-500/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-500/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-500/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-500/trainer_state.json

@@ -0,0 +1,391 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.3013863773357444,
+  "eval_steps": 400,
+  "global_step": 500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006027727546714889,
+      "grad_norm": 6.095676898956299,
+      "learning_rate": 1.0040160642570282e-06,
+      "loss": 1.8585,
+      "step": 10
+    },
+    {
+      "epoch": 0.012055455093429777,
+      "grad_norm": 0.6981305480003357,
+      "learning_rate": 2.0080321285140564e-06,
+      "loss": 1.8607,
+      "step": 20
+    },
+    {
+      "epoch": 0.018083182640144666,
+      "grad_norm": 0.33439213037490845,
+      "learning_rate": 3.0120481927710846e-06,
+      "loss": 1.8035,
+      "step": 30
+    },
+    {
+      "epoch": 0.024110910186859555,
+      "grad_norm": 0.2445673942565918,
+      "learning_rate": 4.016064257028113e-06,
+      "loss": 1.7827,
+      "step": 40
+    },
+    {
+      "epoch": 0.030138637733574444,
+      "grad_norm": 0.22277717292308807,
+      "learning_rate": 5.020080321285141e-06,
+      "loss": 1.7306,
+      "step": 50
+    },
+    {
+      "epoch": 0.03616636528028933,
+      "grad_norm": 0.16803288459777832,
+      "learning_rate": 6.024096385542169e-06,
+      "loss": 1.728,
+      "step": 60
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.1401725709438324,
+      "learning_rate": 7.028112449799197e-06,
+      "loss": 1.6623,
+      "step": 70
+    },
+    {
+      "epoch": 0.04822182037371911,
+      "grad_norm": 0.14054332673549652,
+      "learning_rate": 8.032128514056226e-06,
+      "loss": 1.6226,
+      "step": 80
+    },
+    {
+      "epoch": 0.054249547920433995,
+      "grad_norm": 0.1356767863035202,
+      "learning_rate": 9.036144578313253e-06,
+      "loss": 1.6497,
+      "step": 90
+    },
+    {
+      "epoch": 0.06027727546714889,
+      "grad_norm": 0.14919587969779968,
+      "learning_rate": 1.0040160642570281e-05,
+      "loss": 1.6452,
+      "step": 100
+    },
+    {
+      "epoch": 0.06630500301386377,
+      "grad_norm": 0.16640327870845795,
+      "learning_rate": 1.104417670682731e-05,
+      "loss": 1.6311,
+      "step": 110
+    },
+    {
+      "epoch": 0.07233273056057866,
+      "grad_norm": 0.19770000874996185,
+      "learning_rate": 1.2048192771084338e-05,
+      "loss": 1.5925,
+      "step": 120
+    },
+    {
+      "epoch": 0.07836045810729356,
+      "grad_norm": 0.20399077236652374,
+      "learning_rate": 1.3052208835341367e-05,
+      "loss": 1.5825,
+      "step": 130
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.1512237936258316,
+      "learning_rate": 1.4056224899598394e-05,
+      "loss": 1.5478,
+      "step": 140
+    },
+    {
+      "epoch": 0.09041591320072333,
+      "grad_norm": 0.14794760942459106,
+      "learning_rate": 1.5060240963855424e-05,
+      "loss": 1.5252,
+      "step": 150
+    },
+    {
+      "epoch": 0.09644364074743822,
+      "grad_norm": 0.14903859794139862,
+      "learning_rate": 1.606425702811245e-05,
+      "loss": 1.5427,
+      "step": 160
+    },
+    {
+      "epoch": 0.1024713682941531,
+      "grad_norm": 0.1387884020805359,
+      "learning_rate": 1.706827309236948e-05,
+      "loss": 1.554,
+      "step": 170
+    },
+    {
+      "epoch": 0.10849909584086799,
+      "grad_norm": 0.14016373455524445,
+      "learning_rate": 1.8072289156626505e-05,
+      "loss": 1.5415,
+      "step": 180
+    },
+    {
+      "epoch": 0.11452682338758288,
+      "grad_norm": 0.14568917453289032,
+      "learning_rate": 1.9076305220883535e-05,
+      "loss": 1.5516,
+      "step": 190
+    },
+    {
+      "epoch": 0.12055455093429777,
+      "grad_norm": 0.14068633317947388,
+      "learning_rate": 2.0080321285140562e-05,
+      "loss": 1.5437,
+      "step": 200
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.1428183913230896,
+      "learning_rate": 2.1084337349397593e-05,
+      "loss": 1.5315,
+      "step": 210
+    },
+    {
+      "epoch": 0.13261000602772754,
+      "grad_norm": 0.1416633427143097,
+      "learning_rate": 2.208835341365462e-05,
+      "loss": 1.5276,
+      "step": 220
+    },
+    {
+      "epoch": 0.13863773357444242,
+      "grad_norm": 0.14373181760311127,
+      "learning_rate": 2.309236947791165e-05,
+      "loss": 1.5275,
+      "step": 230
+    },
+    {
+      "epoch": 0.14466546112115733,
+      "grad_norm": 0.16374213993549347,
+      "learning_rate": 2.4096385542168677e-05,
+      "loss": 1.5528,
+      "step": 240
+    },
+    {
+      "epoch": 0.1506931886678722,
+      "grad_norm": 0.13819484412670135,
+      "learning_rate": 2.5100401606425704e-05,
+      "loss": 1.5389,
+      "step": 250
+    },
+    {
+      "epoch": 0.1567209162145871,
+      "grad_norm": 0.14697293937206268,
+      "learning_rate": 2.6104417670682734e-05,
+      "loss": 1.5229,
+      "step": 260
+    },
+    {
+      "epoch": 0.162748643761302,
+      "grad_norm": 0.14389023184776306,
+      "learning_rate": 2.7108433734939758e-05,
+      "loss": 1.5376,
+      "step": 270
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.14889676868915558,
+      "learning_rate": 2.8112449799196788e-05,
+      "loss": 1.5508,
+      "step": 280
+    },
+    {
+      "epoch": 0.17480409885473178,
+      "grad_norm": 0.15245158970355988,
+      "learning_rate": 2.911646586345382e-05,
+      "loss": 1.5342,
+      "step": 290
+    },
+    {
+      "epoch": 0.18083182640144665,
+      "grad_norm": 0.143811896443367,
+      "learning_rate": 3.012048192771085e-05,
+      "loss": 1.5123,
+      "step": 300
+    },
+    {
+      "epoch": 0.18685955394816153,
+      "grad_norm": 0.14010299742221832,
+      "learning_rate": 3.112449799196787e-05,
+      "loss": 1.5355,
+      "step": 310
+    },
+    {
+      "epoch": 0.19288728149487644,
+      "grad_norm": 0.14191724359989166,
+      "learning_rate": 3.21285140562249e-05,
+      "loss": 1.5052,
+      "step": 320
+    },
+    {
+      "epoch": 0.19891500904159132,
+      "grad_norm": 0.13444629311561584,
+      "learning_rate": 3.313253012048193e-05,
+      "loss": 1.5357,
+      "step": 330
+    },
+    {
+      "epoch": 0.2049427365883062,
+      "grad_norm": 0.15248367190361023,
+      "learning_rate": 3.413654618473896e-05,
+      "loss": 1.5171,
+      "step": 340
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.16744668781757355,
+      "learning_rate": 3.5140562248995983e-05,
+      "loss": 1.5379,
+      "step": 350
+    },
+    {
+      "epoch": 0.21699819168173598,
+      "grad_norm": 0.1389068365097046,
+      "learning_rate": 3.614457831325301e-05,
+      "loss": 1.5142,
+      "step": 360
+    },
+    {
+      "epoch": 0.2230259192284509,
+      "grad_norm": 0.13487495481967926,
+      "learning_rate": 3.7148594377510044e-05,
+      "loss": 1.5189,
+      "step": 370
+    },
+    {
+      "epoch": 0.22905364677516576,
+      "grad_norm": 0.13311463594436646,
+      "learning_rate": 3.815261044176707e-05,
+      "loss": 1.5166,
+      "step": 380
+    },
+    {
+      "epoch": 0.23508137432188064,
+      "grad_norm": 0.1321990191936493,
+      "learning_rate": 3.91566265060241e-05,
+      "loss": 1.5218,
+      "step": 390
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "grad_norm": 0.1350693255662918,
+      "learning_rate": 4.0160642570281125e-05,
+      "loss": 1.54,
+      "step": 400
+    },
+    {
+      "epoch": 0.24110910186859555,
+      "eval_loss": 1.4954419136047363,
+      "eval_runtime": 110.0553,
+      "eval_samples_per_second": 9.75,
+      "eval_steps_per_second": 2.444,
+      "step": 400
+    },
+    {
+      "epoch": 0.24713682941531043,
+      "grad_norm": 0.1386946737766266,
+      "learning_rate": 4.116465863453816e-05,
+      "loss": 1.5307,
+      "step": 410
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.1321861445903778,
+      "learning_rate": 4.2168674698795186e-05,
+      "loss": 1.5036,
+      "step": 420
+    },
+    {
+      "epoch": 0.2591922845087402,
+      "grad_norm": 0.13090862333774567,
+      "learning_rate": 4.317269076305221e-05,
+      "loss": 1.5036,
+      "step": 430
+    },
+    {
+      "epoch": 0.2652200120554551,
+      "grad_norm": 0.14390167593955994,
+      "learning_rate": 4.417670682730924e-05,
+      "loss": 1.5245,
+      "step": 440
+    },
+    {
+      "epoch": 0.27124773960216997,
+      "grad_norm": 0.14252203702926636,
+      "learning_rate": 4.5180722891566266e-05,
+      "loss": 1.5361,
+      "step": 450
+    },
+    {
+      "epoch": 0.27727546714888485,
+      "grad_norm": 0.1380358338356018,
+      "learning_rate": 4.61847389558233e-05,
+      "loss": 1.5379,
+      "step": 460
+    },
+    {
+      "epoch": 0.2833031946955998,
+      "grad_norm": 0.17792685329914093,
+      "learning_rate": 4.718875502008032e-05,
+      "loss": 1.5107,
+      "step": 470
+    },
+    {
+      "epoch": 0.28933092224231466,
+      "grad_norm": 0.1495760679244995,
+      "learning_rate": 4.8192771084337354e-05,
+      "loss": 1.5063,
+      "step": 480
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.12926514446735382,
+      "learning_rate": 4.919678714859438e-05,
+      "loss": 1.5149,
+      "step": 490
+    },
+    {
+      "epoch": 0.3013863773357444,
+      "grad_norm": 0.12767350673675537,
+      "learning_rate": 4.999997540156068e-05,
+      "loss": 1.5052,
+      "step": 500
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 4977,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.90727410008064e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:276772c4ef2f26970b87a5557f24d7c42bdfe56a8d7f3d49ad7c2ec614711f92
+size 7032