Upload 18 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/tokenizer.json filter=lfs diff=lfs merge=lfs -text

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/README.md

@@ -0,0 +1,61 @@
+---
+library_name: peft
+license: other
+base_model: /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: sft-qwen2.5-7b-instruct-graph-planning-bs128
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# sft-qwen2.5-7b-instruct-graph-planning-bs128
+
+This model is a fine-tuned version of [/nas/shared/ma4agi/model/Qwen2.5-7B-Instruct](https://huggingface.co//nas/shared/ma4agi/model/Qwen2.5-7B-Instruct) on the graph_planning_train dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 128
+- total_eval_batch_size: 32
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.48.2
+- Pytorch 2.5.1+cu121
+- Datasets 3.2.0
+- Tokenizers 0.21.0

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "model/Qwen2.5-7B-Instruct",
+  "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": 16,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd4e8a00f97b4f8898d8132bf875ce230c5176b5110fc99efcd196b9f515cff8
+size 80792096

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/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
+}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.997242140099283,
+    "num_input_tokens_seen": 69953200,
+    "total_flos": 2.976146663409713e+18,
+    "train_loss": 0.004280612113766934,
+    "train_runtime": 8852.4475,
+    "train_samples_per_second": 4.914,
+    "train_steps_per_second": 0.038
+}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/llamaboard_config.yaml

@@ -0,0 +1,77 @@
+top.booster: auto
+top.checkpoint_path: []
+top.finetuning_type: lora
+top.model_name: Qwen2.5-7B-Instruct
+top.quantization_bit: none
+top.quantization_method: bitsandbytes
+top.rope_scaling: none
+top.template: qwen
+train.additional_target: ''
+train.apollo_rank: 16
+train.apollo_scale: 32
+train.apollo_target: all
+train.apollo_update_interval: 200
+train.badam_mode: layer
+train.badam_switch_interval: 50
+train.badam_switch_mode: ascending
+train.badam_update_ratio: 0.05
+train.batch_size: 2
+train.compute_type: bf16
+train.create_new_adapter: false
+train.cutoff_len: 4096
+train.dataset:
+- graph_planning_train
+train.dataset_dir: data
+train.ds_offload: false
+train.ds_stage: none
+train.extra_args: '{"optim": "adamw_torch"}'
+train.freeze_extra_modules: ''
+train.freeze_trainable_layers: 2
+train.freeze_trainable_modules: all
+train.galore_rank: 16
+train.galore_scale: 2
+train.galore_target: all
+train.galore_update_interval: 200
+train.gradient_accumulation_steps: 16
+train.learning_rate: 1e-4
+train.logging_steps: 1
+train.lora_alpha: 16
+train.lora_dropout: 0
+train.lora_rank: 8
+train.lora_target: ''
+train.loraplus_lr_ratio: 0
+train.lr_scheduler_type: cosine
+train.mask_history: false
+train.max_grad_norm: '1.0'
+train.max_samples: '100000'
+train.neat_packing: false
+train.neftune_alpha: 0
+train.num_train_epochs: '3.0'
+train.packing: false
+train.ppo_score_norm: false
+train.ppo_whiten_rewards: false
+train.pref_beta: 0.1
+train.pref_ftx: 0
+train.pref_loss: sigmoid
+train.report_to:
+- none
+train.resize_vocab: false
+train.reward_model: []
+train.save_steps: 100
+train.swanlab_api_key: ''
+train.swanlab_mode: cloud
+train.swanlab_project: llamafactory
+train.swanlab_run_name: ''
+train.swanlab_workspace: ''
+train.train_on_prompt: false
+train.training_stage: Supervised Fine-Tuning
+train.use_apollo: false
+train.use_badam: false
+train.use_dora: false
+train.use_galore: false
+train.use_llama_pro: false
+train.use_pissa: false
+train.use_rslora: false
+train.use_swanlab: false
+train.val_size: 0
+train.warmup_steps: 0

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/running_log.txt

@@ -0,0 +1,1044 @@
+[INFO|2025-02-10 23:10:14] tokenization_utils_base.py:2032 >> loading file merges.txt
+
+[INFO|2025-02-10 23:10:14] tokenization_utils_base.py:2032 >> loading file tokenizer.json
+
+[INFO|2025-02-10 23:10:14] tokenization_utils_base.py:2032 >> loading file added_tokens.json
+
+[INFO|2025-02-10 23:10:14] tokenization_utils_base.py:2032 >> loading file special_tokens_map.json
+
+[INFO|2025-02-10 23:10:14] tokenization_utils_base.py:2032 >> loading file tokenizer_config.json
+
+[INFO|2025-02-10 23:10:14] tokenization_utils_base.py:2032 >> loading file chat_template.jinja
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2304 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+[INFO|2025-02-10 23:10:15] configuration_utils.py:694 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/config.json
+
+[INFO|2025-02-10 23:10:15] configuration_utils.py:768 >> Model config Qwen2Config {
+  "_name_or_path": "/nas/shared/ma4agi/model/Qwen2.5-7B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2032 >> loading file vocab.json
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2032 >> loading file merges.txt
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2032 >> loading file tokenizer.json
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2032 >> loading file added_tokens.json
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2032 >> loading file special_tokens_map.json
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2032 >> loading file tokenizer_config.json
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2032 >> loading file chat_template.jinja
+
+[INFO|2025-02-10 23:10:15] tokenization_utils_base.py:2304 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+[INFO|2025-02-10 23:10:15] logging.py:157 >> Add <|im_end|> to stop words.
+
+[INFO|2025-02-10 23:10:15] logging.py:157 >> Loading dataset graph_planning/graph_planning_train.json...
+
+[INFO|2025-02-10 23:10:23] configuration_utils.py:694 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/config.json
+
+[INFO|2025-02-10 23:10:23] configuration_utils.py:768 >> Model config Qwen2Config {
+  "_name_or_path": "/nas/shared/ma4agi/model/Qwen2.5-7B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-02-10 23:10:24] modeling_utils.py:3901 >> loading weights file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/model.safetensors.index.json
+
+[INFO|2025-02-10 23:10:24] modeling_utils.py:1582 >> Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
+
+[INFO|2025-02-10 23:10:24] configuration_utils.py:1140 >> Generate config GenerationConfig {
+  "bos_token_id": 151643,
+  "eos_token_id": 151645
+}
+
+
+[INFO|2025-02-10 23:10:27] modeling_utils.py:4888 >> All model checkpoint weights were used when initializing Qwen2ForCausalLM.
+
+
+[INFO|2025-02-10 23:10:27] modeling_utils.py:4896 >> All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
+
+[INFO|2025-02-10 23:10:27] configuration_utils.py:1093 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/generation_config.json
+
+[INFO|2025-02-10 23:10:27] configuration_utils.py:1140 >> Generate config GenerationConfig {
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8
+}
+
+
+[INFO|2025-02-10 23:10:27] logging.py:157 >> Gradient checkpointing enabled.
+
+[INFO|2025-02-10 23:10:27] logging.py:157 >> Using torch SDPA for faster training and inference.
+
+[INFO|2025-02-10 23:10:27] logging.py:157 >> Upcasting trainable params to float32.
+
+[INFO|2025-02-10 23:10:27] logging.py:157 >> Fine-tuning method: LoRA
+
+[INFO|2025-02-10 23:10:27] logging.py:157 >> Found linear modules: o_proj,q_proj,gate_proj,v_proj,up_proj,k_proj,down_proj
+
+[INFO|2025-02-10 23:10:28] logging.py:157 >> trainable params: 20,185,088 || all params: 7,635,801,600 || trainable%: 0.2643
+
+[INFO|2025-02-10 23:10:28] trainer.py:741 >> Using auto half precision backend
+
+[INFO|2025-02-10 23:10:28] trainer.py:2369 >> ***** Running training *****
+
+[INFO|2025-02-10 23:10:28] trainer.py:2370 >>   Num examples = 14,500
+
+[INFO|2025-02-10 23:10:28] trainer.py:2371 >>   Num Epochs = 3
+
+[INFO|2025-02-10 23:10:28] trainer.py:2372 >>   Instantaneous batch size per device = 2
+
+[INFO|2025-02-10 23:10:28] trainer.py:2375 >>   Total train batch size (w. parallel, distributed & accumulation) = 128
+
+[INFO|2025-02-10 23:10:28] trainer.py:2376 >>   Gradient Accumulation steps = 16
+
+[INFO|2025-02-10 23:10:28] trainer.py:2377 >>   Total optimization steps = 339
+
+[INFO|2025-02-10 23:10:28] trainer.py:2378 >>   Number of trainable parameters = 20,185,088
+
+[INFO|2025-02-10 23:10:56] logging.py:157 >> {'loss': 0.2055, 'learning_rate': 9.9998e-05, 'epoch': 0.01, 'throughput': 7338.97}
+
+[INFO|2025-02-10 23:11:22] logging.py:157 >> {'loss': 0.1902, 'learning_rate': 9.9991e-05, 'epoch': 0.02, 'throughput': 7493.38}
+
+[INFO|2025-02-10 23:11:47] logging.py:157 >> {'loss': 0.1421, 'learning_rate': 9.9981e-05, 'epoch': 0.03, 'throughput': 7812.96}
+
+[INFO|2025-02-10 23:12:15] logging.py:157 >> {'loss': 0.1102, 'learning_rate': 9.9966e-05, 'epoch': 0.04, 'throughput': 7668.61}
+
+[INFO|2025-02-10 23:12:40] logging.py:157 >> {'loss': 0.0801, 'learning_rate': 9.9946e-05, 'epoch': 0.04, 'throughput': 7740.76}
+
+[INFO|2025-02-10 23:13:07] logging.py:157 >> {'loss': 0.0574, 'learning_rate': 9.9923e-05, 'epoch': 0.05, 'throughput': 7743.47}
+
+[INFO|2025-02-10 23:13:31] logging.py:157 >> {'loss': 0.0401, 'learning_rate': 9.9895e-05, 'epoch': 0.06, 'throughput': 7830.06}
+
+[INFO|2025-02-10 23:14:00] logging.py:157 >> {'loss': 0.0295, 'learning_rate': 9.9863e-05, 'epoch': 0.07, 'throughput': 7781.82}
+
+[INFO|2025-02-10 23:14:23] logging.py:157 >> {'loss': 0.0262, 'learning_rate': 9.9826e-05, 'epoch': 0.08, 'throughput': 7819.12}
+
+[INFO|2025-02-10 23:14:51] logging.py:157 >> {'loss': 0.0263, 'learning_rate': 9.9785e-05, 'epoch': 0.09, 'throughput': 7807.27}
+
+[INFO|2025-02-10 23:15:15] logging.py:157 >> {'loss': 0.0220, 'learning_rate': 9.9740e-05, 'epoch': 0.10, 'throughput': 7850.91}
+
+[INFO|2025-02-10 23:15:39] logging.py:157 >> {'loss': 0.0202, 'learning_rate': 9.9691e-05, 'epoch': 0.11, 'throughput': 7850.63}
+
+[INFO|2025-02-10 23:16:08] logging.py:157 >> {'loss': 0.0202, 'learning_rate': 9.9638e-05, 'epoch': 0.11, 'throughput': 7823.47}
+
+[INFO|2025-02-10 23:16:38] logging.py:157 >> {'loss': 0.0178, 'learning_rate': 9.9580e-05, 'epoch': 0.12, 'throughput': 7772.01}
+
+[INFO|2025-02-10 23:17:02] logging.py:157 >> {'loss': 0.0158, 'learning_rate': 9.9518e-05, 'epoch': 0.13, 'throughput': 7802.71}
+
+[INFO|2025-02-10 23:17:29] logging.py:157 >> {'loss': 0.0159, 'learning_rate': 9.9451e-05, 'epoch': 0.14, 'throughput': 7786.47}
+
+[INFO|2025-02-10 23:17:56] logging.py:157 >> {'loss': 0.0147, 'learning_rate': 9.9381e-05, 'epoch': 0.15, 'throughput': 7807.80}
+
+[INFO|2025-02-10 23:18:22] logging.py:157 >> {'loss': 0.0138, 'learning_rate': 9.9306e-05, 'epoch': 0.16, 'throughput': 7814.87}
+
+[INFO|2025-02-10 23:18:46] logging.py:157 >> {'loss': 0.0128, 'learning_rate': 9.9227e-05, 'epoch': 0.17, 'throughput': 7835.96}
+
+[INFO|2025-02-10 23:19:13] logging.py:157 >> {'loss': 0.0124, 'learning_rate': 9.9144e-05, 'epoch': 0.18, 'throughput': 7833.94}
+
+[INFO|2025-02-10 23:19:39] logging.py:157 >> {'loss': 0.0116, 'learning_rate': 9.9056e-05, 'epoch': 0.19, 'throughput': 7845.67}
+
+[INFO|2025-02-10 23:20:05] logging.py:157 >> {'loss': 0.0109, 'learning_rate': 9.8964e-05, 'epoch': 0.19, 'throughput': 7844.79}
+
+[INFO|2025-02-10 23:20:32] logging.py:157 >> {'loss': 0.0106, 'learning_rate': 9.8869e-05, 'epoch': 0.20, 'throughput': 7843.06}
+
+[INFO|2025-02-10 23:20:59] logging.py:157 >> {'loss': 0.0099, 'learning_rate': 9.8768e-05, 'epoch': 0.21, 'throughput': 7837.07}
+
+[INFO|2025-02-10 23:21:24] logging.py:157 >> {'loss': 0.0109, 'learning_rate': 9.8664e-05, 'epoch': 0.22, 'throughput': 7848.70}
+
+[INFO|2025-02-10 23:21:48] logging.py:157 >> {'loss': 0.0084, 'learning_rate': 9.8556e-05, 'epoch': 0.23, 'throughput': 7866.16}
+
+[INFO|2025-02-10 23:22:15] logging.py:157 >> {'loss': 0.0081, 'learning_rate': 9.8443e-05, 'epoch': 0.24, 'throughput': 7874.35}
+
+[INFO|2025-02-10 23:22:41] logging.py:157 >> {'loss': 0.0078, 'learning_rate': 9.8326e-05, 'epoch': 0.25, 'throughput': 7877.49}
+
+[INFO|2025-02-10 23:23:09] logging.py:157 >> {'loss': 0.0081, 'learning_rate': 9.8205e-05, 'epoch': 0.26, 'throughput': 7855.15}
+
+[INFO|2025-02-10 23:23:36] logging.py:157 >> {'loss': 0.0087, 'learning_rate': 9.8080e-05, 'epoch': 0.26, 'throughput': 7862.08}
+
+[INFO|2025-02-10 23:24:02] logging.py:157 >> {'loss': 0.0079, 'learning_rate': 9.7951e-05, 'epoch': 0.27, 'throughput': 7859.19}
+
+[INFO|2025-02-10 23:24:28] logging.py:157 >> {'loss': 0.0086, 'learning_rate': 9.7817e-05, 'epoch': 0.28, 'throughput': 7858.66}
+
+[INFO|2025-02-10 23:24:55] logging.py:157 >> {'loss': 0.0079, 'learning_rate': 9.7680e-05, 'epoch': 0.29, 'throughput': 7854.06}
+
+[INFO|2025-02-10 23:25:21] logging.py:157 >> {'loss': 0.0069, 'learning_rate': 9.7538e-05, 'epoch': 0.30, 'throughput': 7844.67}
+
+[INFO|2025-02-10 23:25:48] logging.py:157 >> {'loss': 0.0064, 'learning_rate': 9.7393e-05, 'epoch': 0.31, 'throughput': 7832.92}
+
+[INFO|2025-02-10 23:26:13] logging.py:157 >> {'loss': 0.0060, 'learning_rate': 9.7243e-05, 'epoch': 0.32, 'throughput': 7834.75}
+
+[INFO|2025-02-10 23:26:41] logging.py:157 >> {'loss': 0.0067, 'learning_rate': 9.7089e-05, 'epoch': 0.33, 'throughput': 7845.98}
+
+[INFO|2025-02-10 23:27:07] logging.py:157 >> {'loss': 0.0063, 'learning_rate': 9.6932e-05, 'epoch': 0.34, 'throughput': 7852.67}
+
+[INFO|2025-02-10 23:27:36] logging.py:157 >> {'loss': 0.0058, 'learning_rate': 9.6770e-05, 'epoch': 0.34, 'throughput': 7846.29}
+
+[INFO|2025-02-10 23:28:05] logging.py:157 >> {'loss': 0.0061, 'learning_rate': 9.6604e-05, 'epoch': 0.35, 'throughput': 7836.02}
+
+[INFO|2025-02-10 23:28:29] logging.py:157 >> {'loss': 0.0055, 'learning_rate': 9.6434e-05, 'epoch': 0.36, 'throughput': 7840.93}
+
+[INFO|2025-02-10 23:28:56] logging.py:157 >> {'loss': 0.0050, 'learning_rate': 9.6260e-05, 'epoch': 0.37, 'throughput': 7842.20}
+
+[INFO|2025-02-10 23:29:22] logging.py:157 >> {'loss': 0.0060, 'learning_rate': 9.6082e-05, 'epoch': 0.38, 'throughput': 7833.15}
+
+[INFO|2025-02-10 23:29:48] logging.py:157 >> {'loss': 0.0048, 'learning_rate': 9.5901e-05, 'epoch': 0.39, 'throughput': 7835.19}
+
+[INFO|2025-02-10 23:30:14] logging.py:157 >> {'loss': 0.0047, 'learning_rate': 9.5715e-05, 'epoch': 0.40, 'throughput': 7841.66}
+
+[INFO|2025-02-10 23:30:41] logging.py:157 >> {'loss': 0.0053, 'learning_rate': 9.5525e-05, 'epoch': 0.41, 'throughput': 7848.48}
+
+[INFO|2025-02-10 23:31:05] logging.py:157 >> {'loss': 0.0044, 'learning_rate': 9.5332e-05, 'epoch': 0.41, 'throughput': 7851.65}
+
+[INFO|2025-02-10 23:31:31] logging.py:157 >> {'loss': 0.0043, 'learning_rate': 9.5134e-05, 'epoch': 0.42, 'throughput': 7850.33}
+
+[INFO|2025-02-10 23:31:57] logging.py:157 >> {'loss': 0.0041, 'learning_rate': 9.4933e-05, 'epoch': 0.43, 'throughput': 7853.40}
+
+[INFO|2025-02-10 23:32:26] logging.py:157 >> {'loss': 0.0044, 'learning_rate': 9.4728e-05, 'epoch': 0.44, 'throughput': 7850.51}
+
+[INFO|2025-02-10 23:32:52] logging.py:157 >> {'loss': 0.0040, 'learning_rate': 9.4519e-05, 'epoch': 0.45, 'throughput': 7847.60}
+
+[INFO|2025-02-10 23:33:18] logging.py:157 >> {'loss': 0.0045, 'learning_rate': 9.4306e-05, 'epoch': 0.46, 'throughput': 7853.35}
+
+[INFO|2025-02-10 23:33:43] logging.py:157 >> {'loss': 0.0028, 'learning_rate': 9.4089e-05, 'epoch': 0.47, 'throughput': 7845.79}
+
+[INFO|2025-02-10 23:34:08] logging.py:157 >> {'loss': 0.0034, 'learning_rate': 9.3869e-05, 'epoch': 0.48, 'throughput': 7853.45}
+
+[INFO|2025-02-10 23:34:36] logging.py:157 >> {'loss': 0.0042, 'learning_rate': 9.3645e-05, 'epoch': 0.49, 'throughput': 7852.19}
+
+[INFO|2025-02-10 23:34:59] logging.py:157 >> {'loss': 0.0028, 'learning_rate': 9.3417e-05, 'epoch': 0.49, 'throughput': 7868.08}
+
+[INFO|2025-02-10 23:35:24] logging.py:157 >> {'loss': 0.0035, 'learning_rate': 9.3185e-05, 'epoch': 0.50, 'throughput': 7875.29}
+
+[INFO|2025-02-10 23:35:48] logging.py:157 >> {'loss': 0.0029, 'learning_rate': 9.2950e-05, 'epoch': 0.51, 'throughput': 7884.45}
+
+[INFO|2025-02-10 23:36:13] logging.py:157 >> {'loss': 0.0028, 'learning_rate': 9.2710e-05, 'epoch': 0.52, 'throughput': 7890.10}
+
+[INFO|2025-02-10 23:36:37] logging.py:157 >> {'loss': 0.0027, 'learning_rate': 9.2468e-05, 'epoch': 0.53, 'throughput': 7895.72}
+
+[INFO|2025-02-10 23:37:02] logging.py:157 >> {'loss': 0.0022, 'learning_rate': 9.2221e-05, 'epoch': 0.54, 'throughput': 7896.89}
+
+[INFO|2025-02-10 23:37:34] logging.py:157 >> {'loss': 0.0027, 'learning_rate': 9.1971e-05, 'epoch': 0.55, 'throughput': 7889.78}
+
+[INFO|2025-02-10 23:38:02] logging.py:157 >> {'loss': 0.0025, 'learning_rate': 9.1718e-05, 'epoch': 0.56, 'throughput': 7887.26}
+
+[INFO|2025-02-10 23:38:27] logging.py:157 >> {'loss': 0.0025, 'learning_rate': 9.1461e-05, 'epoch': 0.56, 'throughput': 7894.29}
+
+[INFO|2025-02-10 23:38:57] logging.py:157 >> {'loss': 0.0022, 'learning_rate': 9.1200e-05, 'epoch': 0.57, 'throughput': 7878.69}
+
+[INFO|2025-02-10 23:39:23] logging.py:157 >> {'loss': 0.0023, 'learning_rate': 9.0935e-05, 'epoch': 0.58, 'throughput': 7879.18}
+
+[INFO|2025-02-10 23:39:50] logging.py:157 >> {'loss': 0.0024, 'learning_rate': 9.0668e-05, 'epoch': 0.59, 'throughput': 7877.08}
+
+[INFO|2025-02-10 23:40:18] logging.py:157 >> {'loss': 0.0015, 'learning_rate': 9.0396e-05, 'epoch': 0.60, 'throughput': 7867.14}
+
+[INFO|2025-02-10 23:40:48] logging.py:157 >> {'loss': 0.0033, 'learning_rate': 9.0122e-05, 'epoch': 0.61, 'throughput': 7856.50}
+
+[INFO|2025-02-10 23:41:16] logging.py:157 >> {'loss': 0.0021, 'learning_rate': 8.9843e-05, 'epoch': 0.62, 'throughput': 7860.92}
+
+[INFO|2025-02-10 23:41:41] logging.py:157 >> {'loss': 0.0013, 'learning_rate': 8.9562e-05, 'epoch': 0.63, 'throughput': 7863.29}
+
+[INFO|2025-02-10 23:42:04] logging.py:157 >> {'loss': 0.0016, 'learning_rate': 8.9277e-05, 'epoch': 0.64, 'throughput': 7866.76}
+
+[INFO|2025-02-10 23:42:29] logging.py:157 >> {'loss': 0.0019, 'learning_rate': 8.8988e-05, 'epoch': 0.64, 'throughput': 7870.01}
+
+[INFO|2025-02-10 23:42:56] logging.py:157 >> {'loss': 0.0017, 'learning_rate': 8.8696e-05, 'epoch': 0.65, 'throughput': 7870.22}
+
+[INFO|2025-02-10 23:43:23] logging.py:157 >> {'loss': 0.0015, 'learning_rate': 8.8401e-05, 'epoch': 0.66, 'throughput': 7872.22}
+
+[INFO|2025-02-10 23:43:51] logging.py:157 >> {'loss': 0.0017, 'learning_rate': 8.8103e-05, 'epoch': 0.67, 'throughput': 7866.00}
+
+[INFO|2025-02-10 23:44:17] logging.py:157 >> {'loss': 0.0010, 'learning_rate': 8.7801e-05, 'epoch': 0.68, 'throughput': 7866.60}
+
+[INFO|2025-02-10 23:44:42] logging.py:157 >> {'loss': 0.0018, 'learning_rate': 8.7496e-05, 'epoch': 0.69, 'throughput': 7869.35}
+
+[INFO|2025-02-10 23:45:05] logging.py:157 >> {'loss': 0.0010, 'learning_rate': 8.7188e-05, 'epoch': 0.70, 'throughput': 7871.98}
+
+[INFO|2025-02-10 23:45:31] logging.py:157 >> {'loss': 0.0014, 'learning_rate': 8.6877e-05, 'epoch': 0.71, 'throughput': 7877.09}
+
+[INFO|2025-02-10 23:45:56] logging.py:157 >> {'loss': 0.0010, 'learning_rate': 8.6562e-05, 'epoch': 0.71, 'throughput': 7874.92}
+
+[INFO|2025-02-10 23:46:27] logging.py:157 >> {'loss': 0.0013, 'learning_rate': 8.6245e-05, 'epoch': 0.72, 'throughput': 7863.86}
+
+[INFO|2025-02-10 23:46:50] logging.py:157 >> {'loss': 0.0012, 'learning_rate': 8.5924e-05, 'epoch': 0.73, 'throughput': 7864.79}
+
+[INFO|2025-02-10 23:47:14] logging.py:157 >> {'loss': 0.0013, 'learning_rate': 8.5600e-05, 'epoch': 0.74, 'throughput': 7871.03}
+
+[INFO|2025-02-10 23:47:40] logging.py:157 >> {'loss': 0.0011, 'learning_rate': 8.5273e-05, 'epoch': 0.75, 'throughput': 7869.86}
+
+[INFO|2025-02-10 23:48:09] logging.py:157 >> {'loss': 0.0037, 'learning_rate': 8.4943e-05, 'epoch': 0.76, 'throughput': 7865.40}
+
+[INFO|2025-02-10 23:48:36] logging.py:157 >> {'loss': 0.0007, 'learning_rate': 8.4611e-05, 'epoch': 0.77, 'throughput': 7860.70}
+
+[INFO|2025-02-10 23:49:00] logging.py:157 >> {'loss': 0.0008, 'learning_rate': 8.4275e-05, 'epoch': 0.78, 'throughput': 7865.68}
+
+[INFO|2025-02-10 23:49:28] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 8.3936e-05, 'epoch': 0.79, 'throughput': 7857.75}
+
+[INFO|2025-02-10 23:49:53] logging.py:157 >> {'loss': 0.0010, 'learning_rate': 8.3594e-05, 'epoch': 0.79, 'throughput': 7861.64}
+
+[INFO|2025-02-10 23:50:18] logging.py:157 >> {'loss': 0.0010, 'learning_rate': 8.3249e-05, 'epoch': 0.80, 'throughput': 7864.33}
+
+[INFO|2025-02-10 23:50:43] logging.py:157 >> {'loss': 0.0008, 'learning_rate': 8.2902e-05, 'epoch': 0.81, 'throughput': 7862.08}
+
+[INFO|2025-02-10 23:51:08] logging.py:157 >> {'loss': 0.0014, 'learning_rate': 8.2552e-05, 'epoch': 0.82, 'throughput': 7866.62}
+
+[INFO|2025-02-10 23:51:33] logging.py:157 >> {'loss': 0.0010, 'learning_rate': 8.2199e-05, 'epoch': 0.83, 'throughput': 7869.38}
+
+[INFO|2025-02-10 23:51:59] logging.py:157 >> {'loss': 0.0007, 'learning_rate': 8.1843e-05, 'epoch': 0.84, 'throughput': 7872.25}
+
+[INFO|2025-02-10 23:52:23] logging.py:157 >> {'loss': 0.0008, 'learning_rate': 8.1484e-05, 'epoch': 0.85, 'throughput': 7875.39}
+
+[INFO|2025-02-10 23:52:51] logging.py:157 >> {'loss': 0.0011, 'learning_rate': 8.1123e-05, 'epoch': 0.86, 'throughput': 7868.64}
+
+[INFO|2025-02-10 23:53:21] logging.py:157 >> {'loss': 0.0009, 'learning_rate': 8.0759e-05, 'epoch': 0.86, 'throughput': 7865.85}
+
+[INFO|2025-02-10 23:53:44] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 8.0392e-05, 'epoch': 0.87, 'throughput': 7870.61}
+
+[INFO|2025-02-10 23:54:08] logging.py:157 >> {'loss': 0.0009, 'learning_rate': 8.0023e-05, 'epoch': 0.88, 'throughput': 7876.56}
+
+[INFO|2025-02-10 23:54:08] trainer.py:3910 >> Saving model checkpoint to saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-100
+
+[INFO|2025-02-10 23:54:08] configuration_utils.py:694 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/config.json
+
+[INFO|2025-02-10 23:54:08] configuration_utils.py:768 >> Model config Qwen2Config {
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-02-10 23:54:08] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-100/tokenizer_config.json
+
+[INFO|2025-02-10 23:54:08] tokenization_utils_base.py:2500 >> Special tokens file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-100/special_tokens_map.json
+
+[INFO|2025-02-10 23:54:33] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 7.9651e-05, 'epoch': 0.89, 'throughput': 7876.49}
+
+[INFO|2025-02-10 23:55:01] logging.py:157 >> {'loss': 0.0008, 'learning_rate': 7.9277e-05, 'epoch': 0.90, 'throughput': 7875.34}
+
+[INFO|2025-02-10 23:55:25] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 7.8900e-05, 'epoch': 0.91, 'throughput': 7878.05}
+
+[INFO|2025-02-10 23:55:48] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 7.8520e-05, 'epoch': 0.92, 'throughput': 7881.27}
+
+[INFO|2025-02-10 23:56:16] logging.py:157 >> {'loss': 0.0008, 'learning_rate': 7.8139e-05, 'epoch': 0.93, 'throughput': 7877.02}
+
+[INFO|2025-02-10 23:56:44] logging.py:157 >> {'loss': 0.0007, 'learning_rate': 7.7754e-05, 'epoch': 0.94, 'throughput': 7877.23}
+
+[INFO|2025-02-10 23:57:08] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 7.7368e-05, 'epoch': 0.94, 'throughput': 7878.63}
+
+[INFO|2025-02-10 23:57:34] logging.py:157 >> {'loss': 0.0012, 'learning_rate': 7.6979e-05, 'epoch': 0.95, 'throughput': 7883.06}
+
+[INFO|2025-02-10 23:57:59] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 7.6588e-05, 'epoch': 0.96, 'throughput': 7881.65}
+
+[INFO|2025-02-10 23:58:23] logging.py:157 >> {'loss': 0.0008, 'learning_rate': 7.6194e-05, 'epoch': 0.97, 'throughput': 7883.60}
+
+[INFO|2025-02-10 23:58:51] logging.py:157 >> {'loss': 0.0007, 'learning_rate': 7.5798e-05, 'epoch': 0.98, 'throughput': 7884.35}
+
+[INFO|2025-02-10 23:59:14] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 7.5400e-05, 'epoch': 0.99, 'throughput': 7889.50}
+
+[INFO|2025-02-10 23:59:42] logging.py:157 >> {'loss': 0.0007, 'learning_rate': 7.5000e-05, 'epoch': 1.00, 'throughput': 7889.54}
+
+[INFO|2025-02-11 00:00:15] logging.py:157 >> {'loss': 0.0014, 'learning_rate': 7.4598e-05, 'epoch': 1.01, 'throughput': 7889.22}
+
+[INFO|2025-02-11 00:00:42] logging.py:157 >> {'loss': 0.0007, 'learning_rate': 7.4193e-05, 'epoch': 1.02, 'throughput': 7886.62}
+
+[INFO|2025-02-11 00:01:08] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 7.3787e-05, 'epoch': 1.03, 'throughput': 7886.10}
+
+[INFO|2025-02-11 00:01:33] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 7.3378e-05, 'epoch': 1.04, 'throughput': 7886.46}
+
+[INFO|2025-02-11 00:01:56] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 7.2967e-05, 'epoch': 1.04, 'throughput': 7889.90}
+
+[INFO|2025-02-11 00:02:21] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 7.2555e-05, 'epoch': 1.05, 'throughput': 7889.45}
+
+[INFO|2025-02-11 00:02:50] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 7.2140e-05, 'epoch': 1.06, 'throughput': 7886.78}
+
+[INFO|2025-02-11 00:03:14] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 7.1724e-05, 'epoch': 1.07, 'throughput': 7893.28}
+
+[INFO|2025-02-11 00:03:40] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 7.1306e-05, 'epoch': 1.08, 'throughput': 7896.08}
+
+[INFO|2025-02-11 00:04:05] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 7.0886e-05, 'epoch': 1.09, 'throughput': 7901.06}
+
+[INFO|2025-02-11 00:04:28] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 7.0464e-05, 'epoch': 1.10, 'throughput': 7904.61}
+
+[INFO|2025-02-11 00:04:54] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 7.0040e-05, 'epoch': 1.11, 'throughput': 7904.82}
+
+[INFO|2025-02-11 00:05:22] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 6.9615e-05, 'epoch': 1.11, 'throughput': 7905.63}
+
+[INFO|2025-02-11 00:05:49] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 6.9188e-05, 'epoch': 1.12, 'throughput': 7904.23}
+
+[INFO|2025-02-11 00:06:16] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 6.8759e-05, 'epoch': 1.13, 'throughput': 7903.09}
+
+[INFO|2025-02-11 00:06:46] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 6.8329e-05, 'epoch': 1.14, 'throughput': 7898.37}
+
+[INFO|2025-02-11 00:07:13] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 6.7897e-05, 'epoch': 1.15, 'throughput': 7897.68}
+
+[INFO|2025-02-11 00:07:41] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 6.7463e-05, 'epoch': 1.16, 'throughput': 7897.31}
+
+[INFO|2025-02-11 00:08:08] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 6.7028e-05, 'epoch': 1.17, 'throughput': 7895.24}
+
+[INFO|2025-02-11 00:08:34] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 6.6592e-05, 'epoch': 1.18, 'throughput': 7895.50}
+
+[INFO|2025-02-11 00:09:01] logging.py:157 >> {'loss': 0.0009, 'learning_rate': 6.6154e-05, 'epoch': 1.19, 'throughput': 7896.09}
+
+[INFO|2025-02-11 00:09:29] logging.py:157 >> {'loss': 0.0014, 'learning_rate': 6.5715e-05, 'epoch': 1.19, 'throughput': 7892.01}
+
+[INFO|2025-02-11 00:09:54] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 6.5274e-05, 'epoch': 1.20, 'throughput': 7891.89}
+
+[INFO|2025-02-11 00:10:22] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 6.4833e-05, 'epoch': 1.21, 'throughput': 7888.92}
+
+[INFO|2025-02-11 00:10:49] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 6.4389e-05, 'epoch': 1.22, 'throughput': 7888.52}
+
+[INFO|2025-02-11 00:11:14] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 6.3945e-05, 'epoch': 1.23, 'throughput': 7890.66}
+
+[INFO|2025-02-11 00:11:36] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 6.3500e-05, 'epoch': 1.24, 'throughput': 7892.64}
+
+[INFO|2025-02-11 00:12:00] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 6.3053e-05, 'epoch': 1.25, 'throughput': 7894.92}
+
+[INFO|2025-02-11 00:12:29] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 6.2605e-05, 'epoch': 1.26, 'throughput': 7892.49}
+
+[INFO|2025-02-11 00:12:54] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 6.2156e-05, 'epoch': 1.26, 'throughput': 7894.05}
+
+[INFO|2025-02-11 00:13:20] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 6.1706e-05, 'epoch': 1.27, 'throughput': 7893.70}
+
+[INFO|2025-02-11 00:13:42] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 6.1255e-05, 'epoch': 1.28, 'throughput': 7896.62}
+
+[INFO|2025-02-11 00:14:09] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 6.0803e-05, 'epoch': 1.29, 'throughput': 7895.09}
+
+[INFO|2025-02-11 00:14:34] logging.py:157 >> {'loss': 0.0005, 'learning_rate': 6.0350e-05, 'epoch': 1.30, 'throughput': 7897.58}
+
+[INFO|2025-02-11 00:15:00] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.9896e-05, 'epoch': 1.31, 'throughput': 7896.39}
+
+[INFO|2025-02-11 00:15:25] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 5.9442e-05, 'epoch': 1.32, 'throughput': 7897.88}
+
+[INFO|2025-02-11 00:15:48] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 5.8986e-05, 'epoch': 1.33, 'throughput': 7902.00}
+
+[INFO|2025-02-11 00:16:14] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 5.8530e-05, 'epoch': 1.34, 'throughput': 7900.57}
+
+[INFO|2025-02-11 00:16:41] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.8073e-05, 'epoch': 1.34, 'throughput': 7901.36}
+
+[INFO|2025-02-11 00:17:06] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.7616e-05, 'epoch': 1.35, 'throughput': 7898.31}
+
+[INFO|2025-02-11 00:17:30] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.7157e-05, 'epoch': 1.36, 'throughput': 7901.98}
+
+[INFO|2025-02-11 00:17:58] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 5.6699e-05, 'epoch': 1.37, 'throughput': 7898.91}
+
+[INFO|2025-02-11 00:18:22] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 5.6239e-05, 'epoch': 1.38, 'throughput': 7901.45}
+
+[INFO|2025-02-11 00:18:47] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 5.5779e-05, 'epoch': 1.39, 'throughput': 7900.71}
+
+[INFO|2025-02-11 00:19:14] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 5.5319e-05, 'epoch': 1.40, 'throughput': 7898.98}
+
+[INFO|2025-02-11 00:19:41] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.4858e-05, 'epoch': 1.41, 'throughput': 7897.63}
+
+[INFO|2025-02-11 00:20:08] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.4396e-05, 'epoch': 1.41, 'throughput': 7897.71}
+
+[INFO|2025-02-11 00:20:33] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.3935e-05, 'epoch': 1.42, 'throughput': 7897.58}
+
+[INFO|2025-02-11 00:20:57] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 5.3472e-05, 'epoch': 1.43, 'throughput': 7900.98}
+
+[INFO|2025-02-11 00:21:26] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.3010e-05, 'epoch': 1.44, 'throughput': 7899.49}
+
+[INFO|2025-02-11 00:21:50] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 5.2547e-05, 'epoch': 1.45, 'throughput': 7899.53}
+
+[INFO|2025-02-11 00:22:13] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.2085e-05, 'epoch': 1.46, 'throughput': 7901.89}
+
+[INFO|2025-02-11 00:22:39] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 5.1621e-05, 'epoch': 1.47, 'throughput': 7902.03}
+
+[INFO|2025-02-11 00:23:05] logging.py:157 >> {'loss': 0.0016, 'learning_rate': 5.1158e-05, 'epoch': 1.48, 'throughput': 7901.05}
+
+[INFO|2025-02-11 00:23:32] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 5.0695e-05, 'epoch': 1.49, 'throughput': 7901.53}
+
+[INFO|2025-02-11 00:23:57] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.0232e-05, 'epoch': 1.49, 'throughput': 7901.79}
+
+[INFO|2025-02-11 00:24:22] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.9768e-05, 'epoch': 1.50, 'throughput': 7901.03}
+
+[INFO|2025-02-11 00:24:45] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 4.9305e-05, 'epoch': 1.51, 'throughput': 7903.96}
+
+[INFO|2025-02-11 00:25:11] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.8842e-05, 'epoch': 1.52, 'throughput': 7905.12}
+
+[INFO|2025-02-11 00:25:33] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.8379e-05, 'epoch': 1.53, 'throughput': 7909.17}
+
+[INFO|2025-02-11 00:25:58] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.7915e-05, 'epoch': 1.54, 'throughput': 7909.02}
+
+[INFO|2025-02-11 00:26:29] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 4.7453e-05, 'epoch': 1.55, 'throughput': 7905.59}
+
+[INFO|2025-02-11 00:26:56] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.6990e-05, 'epoch': 1.56, 'throughput': 7905.47}
+
+[INFO|2025-02-11 00:27:22] logging.py:157 >> {'loss': 0.0027, 'learning_rate': 4.6528e-05, 'epoch': 1.56, 'throughput': 7904.70}
+
+[INFO|2025-02-11 00:27:47] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 4.6065e-05, 'epoch': 1.57, 'throughput': 7905.80}
+
+[INFO|2025-02-11 00:28:12] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 4.5604e-05, 'epoch': 1.58, 'throughput': 7907.65}
+
+[INFO|2025-02-11 00:28:39] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 4.5142e-05, 'epoch': 1.59, 'throughput': 7907.43}
+
+[INFO|2025-02-11 00:29:09] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.4681e-05, 'epoch': 1.60, 'throughput': 7902.86}
+
+[INFO|2025-02-11 00:29:35] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 4.4221e-05, 'epoch': 1.61, 'throughput': 7903.10}
+
+[INFO|2025-02-11 00:30:02] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 4.3761e-05, 'epoch': 1.62, 'throughput': 7904.08}
+
+[INFO|2025-02-11 00:30:28] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 4.3301e-05, 'epoch': 1.63, 'throughput': 7905.25}
+
+[INFO|2025-02-11 00:30:50] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 4.2843e-05, 'epoch': 1.64, 'throughput': 7908.78}
+
+[INFO|2025-02-11 00:31:13] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.2384e-05, 'epoch': 1.64, 'throughput': 7912.45}
+
+[INFO|2025-02-11 00:31:40] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 4.1927e-05, 'epoch': 1.65, 'throughput': 7912.95}
+
+[INFO|2025-02-11 00:32:06] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 4.1470e-05, 'epoch': 1.66, 'throughput': 7913.14}
+
+[INFO|2025-02-11 00:32:33] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.1014e-05, 'epoch': 1.67, 'throughput': 7914.08}
+
+[INFO|2025-02-11 00:33:02] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 4.0558e-05, 'epoch': 1.68, 'throughput': 7911.16}
+
+[INFO|2025-02-11 00:33:29] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 4.0104e-05, 'epoch': 1.69, 'throughput': 7911.74}
+
+[INFO|2025-02-11 00:33:56] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 3.9650e-05, 'epoch': 1.70, 'throughput': 7911.05}
+
+[INFO|2025-02-11 00:34:21] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.9197e-05, 'epoch': 1.71, 'throughput': 7910.45}
+
+[INFO|2025-02-11 00:34:46] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 3.8745e-05, 'epoch': 1.71, 'throughput': 7911.47}
+
+[INFO|2025-02-11 00:35:10] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.8294e-05, 'epoch': 1.72, 'throughput': 7913.07}
+
+[INFO|2025-02-11 00:35:36] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.7844e-05, 'epoch': 1.73, 'throughput': 7912.61}
+
+[INFO|2025-02-11 00:36:01] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.7395e-05, 'epoch': 1.74, 'throughput': 7914.15}
+
+[INFO|2025-02-11 00:36:33] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 3.6947e-05, 'epoch': 1.75, 'throughput': 7908.90}
+
+[INFO|2025-02-11 00:36:56] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.6500e-05, 'epoch': 1.76, 'throughput': 7909.88}
+
+[INFO|2025-02-11 00:37:27] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 3.6055e-05, 'epoch': 1.77, 'throughput': 7904.23}
+
+[INFO|2025-02-11 00:37:27] trainer.py:3910 >> Saving model checkpoint to saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-200
+
+[INFO|2025-02-11 00:37:28] configuration_utils.py:694 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/config.json
+
+[INFO|2025-02-11 00:37:28] configuration_utils.py:768 >> Model config Qwen2Config {
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-02-11 00:37:28] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-200/tokenizer_config.json
+
+[INFO|2025-02-11 00:37:28] tokenization_utils_base.py:2500 >> Special tokens file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-200/special_tokens_map.json
+
+[INFO|2025-02-11 00:37:58] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.5611e-05, 'epoch': 1.78, 'throughput': 7898.91}
+
+[INFO|2025-02-11 00:38:25] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.5167e-05, 'epoch': 1.79, 'throughput': 7897.44}
+
+[INFO|2025-02-11 00:38:53] logging.py:157 >> {'loss': 0.0013, 'learning_rate': 3.4726e-05, 'epoch': 1.79, 'throughput': 7894.63}
+
+[INFO|2025-02-11 00:39:20] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.4285e-05, 'epoch': 1.80, 'throughput': 7893.32}
+
+[INFO|2025-02-11 00:39:45] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.3846e-05, 'epoch': 1.81, 'throughput': 7893.48}
+
+[INFO|2025-02-11 00:40:10] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 3.3408e-05, 'epoch': 1.82, 'throughput': 7897.32}
+
+[INFO|2025-02-11 00:40:34] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.2972e-05, 'epoch': 1.83, 'throughput': 7898.07}
+
+[INFO|2025-02-11 00:41:01] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 3.2537e-05, 'epoch': 1.84, 'throughput': 7898.41}
+
+[INFO|2025-02-11 00:41:27] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.2103e-05, 'epoch': 1.85, 'throughput': 7898.41}
+
+[INFO|2025-02-11 00:41:56] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 3.1671e-05, 'epoch': 1.86, 'throughput': 7897.79}
+
+[INFO|2025-02-11 00:42:22] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.1241e-05, 'epoch': 1.86, 'throughput': 7896.37}
+
+[INFO|2025-02-11 00:42:45] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 3.0812e-05, 'epoch': 1.87, 'throughput': 7897.48}
+
+[INFO|2025-02-11 00:43:13] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.0385e-05, 'epoch': 1.88, 'throughput': 7898.39}
+
+[INFO|2025-02-11 00:43:39] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.9960e-05, 'epoch': 1.89, 'throughput': 7897.60}
+
+[INFO|2025-02-11 00:44:03] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.9536e-05, 'epoch': 1.90, 'throughput': 7898.54}
+
+[INFO|2025-02-11 00:44:28] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.9114e-05, 'epoch': 1.91, 'throughput': 7900.99}
+
+[INFO|2025-02-11 00:44:55] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.8694e-05, 'epoch': 1.92, 'throughput': 7899.67}
+
+[INFO|2025-02-11 00:45:23] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.8276e-05, 'epoch': 1.93, 'throughput': 7898.82}
+
+[INFO|2025-02-11 00:45:50] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.7860e-05, 'epoch': 1.94, 'throughput': 7898.40}
+
+[INFO|2025-02-11 00:46:19] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.7445e-05, 'epoch': 1.94, 'throughput': 7895.44}
+
+[INFO|2025-02-11 00:46:46] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 2.7033e-05, 'epoch': 1.95, 'throughput': 7892.18}
+
+[INFO|2025-02-11 00:47:13] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.6622e-05, 'epoch': 1.96, 'throughput': 7890.65}
+
+[INFO|2025-02-11 00:47:36] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.6213e-05, 'epoch': 1.97, 'throughput': 7892.67}
+
+[INFO|2025-02-11 00:48:03] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.5807e-05, 'epoch': 1.98, 'throughput': 7892.10}
+
+[INFO|2025-02-11 00:48:31] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.5402e-05, 'epoch': 1.99, 'throughput': 7891.47}
+
+[INFO|2025-02-11 00:48:57] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 2.5000e-05, 'epoch': 2.00, 'throughput': 7891.65}
+
+[INFO|2025-02-11 00:49:32] logging.py:157 >> {'loss': 0.0011, 'learning_rate': 2.4600e-05, 'epoch': 2.01, 'throughput': 7890.98}
+
+[INFO|2025-02-11 00:49:59] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.4202e-05, 'epoch': 2.02, 'throughput': 7888.59}
+
+[INFO|2025-02-11 00:50:25] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.3806e-05, 'epoch': 2.03, 'throughput': 7889.79}
+
+[INFO|2025-02-11 00:50:54] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 2.3412e-05, 'epoch': 2.04, 'throughput': 7887.38}
+
+[INFO|2025-02-11 00:51:17] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.3021e-05, 'epoch': 2.04, 'throughput': 7889.45}
+
+[INFO|2025-02-11 00:51:46] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.2632e-05, 'epoch': 2.05, 'throughput': 7887.07}
+
+[INFO|2025-02-11 00:52:11] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.2246e-05, 'epoch': 2.06, 'throughput': 7889.59}
+
+[INFO|2025-02-11 00:52:43] logging.py:157 >> {'loss': 0.0025, 'learning_rate': 2.1861e-05, 'epoch': 2.07, 'throughput': 7885.55}
+
+[INFO|2025-02-11 00:53:13] logging.py:157 >> {'loss': 0.0013, 'learning_rate': 2.1480e-05, 'epoch': 2.08, 'throughput': 7881.47}
+
+[INFO|2025-02-11 00:53:37] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.1100e-05, 'epoch': 2.09, 'throughput': 7884.07}
+
+[INFO|2025-02-11 00:54:04] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.0723e-05, 'epoch': 2.10, 'throughput': 7884.69}
+
+[INFO|2025-02-11 00:54:29] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.0349e-05, 'epoch': 2.11, 'throughput': 7886.69}
+
+[INFO|2025-02-11 00:54:55] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.9977e-05, 'epoch': 2.11, 'throughput': 7886.90}
+
+[INFO|2025-02-11 00:55:19] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.9608e-05, 'epoch': 2.12, 'throughput': 7888.14}
+
+[INFO|2025-02-11 00:55:44] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.9241e-05, 'epoch': 2.13, 'throughput': 7889.13}
+
+[INFO|2025-02-11 00:56:09] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.8877e-05, 'epoch': 2.14, 'throughput': 7888.30}
+
+[INFO|2025-02-11 00:56:35] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.8516e-05, 'epoch': 2.15, 'throughput': 7887.77}
+
+[INFO|2025-02-11 00:56:59] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.8157e-05, 'epoch': 2.16, 'throughput': 7889.80}
+
+[INFO|2025-02-11 00:57:28] logging.py:157 >> {'loss': 0.0006, 'learning_rate': 1.7801e-05, 'epoch': 2.17, 'throughput': 7887.66}
+
+[INFO|2025-02-11 00:57:55] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.7448e-05, 'epoch': 2.18, 'throughput': 7886.91}
+
+[INFO|2025-02-11 00:58:18] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.7098e-05, 'epoch': 2.19, 'throughput': 7889.08}
+
+[INFO|2025-02-11 00:58:42] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.6751e-05, 'epoch': 2.19, 'throughput': 7890.61}
+
+[INFO|2025-02-11 00:59:08] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.6406e-05, 'epoch': 2.20, 'throughput': 7889.22}
+
+[INFO|2025-02-11 00:59:33] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.6064e-05, 'epoch': 2.21, 'throughput': 7890.88}
+
+[INFO|2025-02-11 00:59:58] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.5725e-05, 'epoch': 2.22, 'throughput': 7890.72}
+
+[INFO|2025-02-11 01:00:24] logging.py:157 >> {'loss': 0.0012, 'learning_rate': 1.5389e-05, 'epoch': 2.23, 'throughput': 7890.97}
+
+[INFO|2025-02-11 01:00:47] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.5057e-05, 'epoch': 2.24, 'throughput': 7892.37}
+
+[INFO|2025-02-11 01:01:11] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 1.4727e-05, 'epoch': 2.25, 'throughput': 7894.16}
+
+[INFO|2025-02-11 01:01:34] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.4400e-05, 'epoch': 2.26, 'throughput': 7894.39}
+
+[INFO|2025-02-11 01:02:01] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.4076e-05, 'epoch': 2.26, 'throughput': 7895.04}
+
+[INFO|2025-02-11 01:02:29] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.3755e-05, 'epoch': 2.27, 'throughput': 7894.62}
+
+[INFO|2025-02-11 01:02:57] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.3438e-05, 'epoch': 2.28, 'throughput': 7893.02}
+
+[INFO|2025-02-11 01:03:28] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.3123e-05, 'epoch': 2.29, 'throughput': 7891.68}
+
+[INFO|2025-02-11 01:03:52] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.2812e-05, 'epoch': 2.30, 'throughput': 7892.09}
+
+[INFO|2025-02-11 01:04:18] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.2504e-05, 'epoch': 2.31, 'throughput': 7894.29}
+
+[INFO|2025-02-11 01:04:46] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.2199e-05, 'epoch': 2.32, 'throughput': 7893.79}
+
+[INFO|2025-02-11 01:05:13] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.1897e-05, 'epoch': 2.33, 'throughput': 7894.54}
+
+[INFO|2025-02-11 01:05:37] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.1599e-05, 'epoch': 2.34, 'throughput': 7894.72}
+
+[INFO|2025-02-11 01:06:03] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.1304e-05, 'epoch': 2.34, 'throughput': 7893.77}
+
+[INFO|2025-02-11 01:06:29] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.1012e-05, 'epoch': 2.35, 'throughput': 7895.00}
+
+[INFO|2025-02-11 01:06:55] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.0723e-05, 'epoch': 2.36, 'throughput': 7895.05}
+
+[INFO|2025-02-11 01:07:22] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.0438e-05, 'epoch': 2.37, 'throughput': 7895.15}
+
+[INFO|2025-02-11 01:07:48] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.0157e-05, 'epoch': 2.38, 'throughput': 7895.90}
+
+[INFO|2025-02-11 01:08:14] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 9.8785e-06, 'epoch': 2.39, 'throughput': 7897.00}
+
+[INFO|2025-02-11 01:08:43] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 9.6037e-06, 'epoch': 2.40, 'throughput': 7893.16}
+
+[INFO|2025-02-11 01:09:10] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 9.3324e-06, 'epoch': 2.41, 'throughput': 7891.24}
+
+[INFO|2025-02-11 01:09:34] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 9.0646e-06, 'epoch': 2.41, 'throughput': 7893.30}
+
+[INFO|2025-02-11 01:10:00] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 8.8003e-06, 'epoch': 2.42, 'throughput': 7893.02}
+
+[INFO|2025-02-11 01:10:25] logging.py:157 >> {'loss': 0.0000, 'learning_rate': 8.5395e-06, 'epoch': 2.43, 'throughput': 7892.52}
+
+[INFO|2025-02-11 01:10:51] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 8.2823e-06, 'epoch': 2.44, 'throughput': 7892.63}
+
+[INFO|2025-02-11 01:11:17] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 8.0287e-06, 'epoch': 2.45, 'throughput': 7892.41}
+
+[INFO|2025-02-11 01:11:45] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 7.7786e-06, 'epoch': 2.46, 'throughput': 7889.96}
+
+[INFO|2025-02-11 01:12:10] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 7.5322e-06, 'epoch': 2.47, 'throughput': 7892.18}
+
+[INFO|2025-02-11 01:12:34] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 7.2895e-06, 'epoch': 2.48, 'throughput': 7893.12}
+
+[INFO|2025-02-11 01:13:03] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 7.0504e-06, 'epoch': 2.49, 'throughput': 7891.37}
+
+[INFO|2025-02-11 01:13:30] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 6.8150e-06, 'epoch': 2.49, 'throughput': 7891.23}
+
+[INFO|2025-02-11 01:13:58] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 6.5834e-06, 'epoch': 2.50, 'throughput': 7890.21}
+
+[INFO|2025-02-11 01:14:25] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 6.3554e-06, 'epoch': 2.51, 'throughput': 7889.96}
+
+[INFO|2025-02-11 01:14:53] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 6.1312e-06, 'epoch': 2.52, 'throughput': 7889.86}
+
+[INFO|2025-02-11 01:15:20] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 5.9108e-06, 'epoch': 2.53, 'throughput': 7889.91}
+
+[INFO|2025-02-11 01:15:44] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 5.6941e-06, 'epoch': 2.54, 'throughput': 7891.05}
+
+[INFO|2025-02-11 01:16:10] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 5.4813e-06, 'epoch': 2.55, 'throughput': 7890.97}
+
+[INFO|2025-02-11 01:16:34] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 5.2723e-06, 'epoch': 2.56, 'throughput': 7891.95}
+
+[INFO|2025-02-11 01:16:58] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 5.0671e-06, 'epoch': 2.56, 'throughput': 7893.32}
+
+[INFO|2025-02-11 01:17:24] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.8657e-06, 'epoch': 2.57, 'throughput': 7893.98}
+
+[INFO|2025-02-11 01:17:51] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 4.6683e-06, 'epoch': 2.58, 'throughput': 7892.90}
+
+[INFO|2025-02-11 01:18:18] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 4.4748e-06, 'epoch': 2.59, 'throughput': 7892.36}
+
+[INFO|2025-02-11 01:18:46] logging.py:157 >> {'loss': 0.0010, 'learning_rate': 4.2851e-06, 'epoch': 2.60, 'throughput': 7891.40}
+
+[INFO|2025-02-11 01:19:11] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 4.0994e-06, 'epoch': 2.61, 'throughput': 7891.16}
+
+[INFO|2025-02-11 01:19:35] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.9176e-06, 'epoch': 2.62, 'throughput': 7892.02}
+
+[INFO|2025-02-11 01:19:59] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.7398e-06, 'epoch': 2.63, 'throughput': 7892.85}
+
+[INFO|2025-02-11 01:20:26] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.5660e-06, 'epoch': 2.64, 'throughput': 7892.72}
+
+[INFO|2025-02-11 01:20:52] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 3.3961e-06, 'epoch': 2.64, 'throughput': 7893.04}
+
+[INFO|2025-02-11 01:21:17] logging.py:157 >> {'loss': 0.0000, 'learning_rate': 3.2303e-06, 'epoch': 2.65, 'throughput': 7892.55}
+
+[INFO|2025-02-11 01:21:17] trainer.py:3910 >> Saving model checkpoint to saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-300
+
+[INFO|2025-02-11 01:21:17] configuration_utils.py:694 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/config.json
+
+[INFO|2025-02-11 01:21:17] configuration_utils.py:768 >> Model config Qwen2Config {
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-02-11 01:21:17] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-300/tokenizer_config.json
+
+[INFO|2025-02-11 01:21:17] tokenization_utils_base.py:2500 >> Special tokens file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-300/special_tokens_map.json
+
+[INFO|2025-02-11 01:21:42] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.0684e-06, 'epoch': 2.66, 'throughput': 7893.70}
+
+[INFO|2025-02-11 01:22:07] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.9106e-06, 'epoch': 2.67, 'throughput': 7893.69}
+
+[INFO|2025-02-11 01:22:31] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.7569e-06, 'epoch': 2.68, 'throughput': 7893.79}
+
+[INFO|2025-02-11 01:22:58] logging.py:157 >> {'loss': 0.0004, 'learning_rate': 2.6071e-06, 'epoch': 2.69, 'throughput': 7893.85}
+
+[INFO|2025-02-11 01:23:27] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.4615e-06, 'epoch': 2.70, 'throughput': 7891.77}
+
+[INFO|2025-02-11 01:23:51] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.3200e-06, 'epoch': 2.71, 'throughput': 7892.22}
+
+[INFO|2025-02-11 01:24:18] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.1825e-06, 'epoch': 2.71, 'throughput': 7891.29}
+
+[INFO|2025-02-11 01:24:43] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.0492e-06, 'epoch': 2.72, 'throughput': 7892.66}
+
+[INFO|2025-02-11 01:25:08] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.9199e-06, 'epoch': 2.73, 'throughput': 7892.67}
+
+[INFO|2025-02-11 01:25:34] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.7948e-06, 'epoch': 2.74, 'throughput': 7892.91}
+
+[INFO|2025-02-11 01:26:01] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.6739e-06, 'epoch': 2.75, 'throughput': 7892.87}
+
+[INFO|2025-02-11 01:26:28] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.5570e-06, 'epoch': 2.76, 'throughput': 7892.66}
+
+[INFO|2025-02-11 01:26:55] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.4444e-06, 'epoch': 2.77, 'throughput': 7894.09}
+
+[INFO|2025-02-11 01:27:20] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.3359e-06, 'epoch': 2.78, 'throughput': 7894.56}
+
+[INFO|2025-02-11 01:27:46] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 1.2316e-06, 'epoch': 2.79, 'throughput': 7893.92}
+
+[INFO|2025-02-11 01:28:12] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.1315e-06, 'epoch': 2.79, 'throughput': 7894.65}
+
+[INFO|2025-02-11 01:28:38] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.0356e-06, 'epoch': 2.80, 'throughput': 7893.46}
+
+[INFO|2025-02-11 01:29:02] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 9.4386e-07, 'epoch': 2.81, 'throughput': 7894.99}
+
+[INFO|2025-02-11 01:29:31] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 8.5636e-07, 'epoch': 2.82, 'throughput': 7894.00}
+
+[INFO|2025-02-11 01:29:57] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 7.7308e-07, 'epoch': 2.83, 'throughput': 7895.06}
+
+[INFO|2025-02-11 01:30:24] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 6.9403e-07, 'epoch': 2.84, 'throughput': 7894.58}
+
+[INFO|2025-02-11 01:30:53] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 6.1921e-07, 'epoch': 2.85, 'throughput': 7893.93}
+
+[INFO|2025-02-11 01:31:19] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 5.4864e-07, 'epoch': 2.86, 'throughput': 7893.63}
+
+[INFO|2025-02-11 01:31:45] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 4.8231e-07, 'epoch': 2.86, 'throughput': 7893.28}
+
+[INFO|2025-02-11 01:32:10] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 4.2023e-07, 'epoch': 2.87, 'throughput': 7894.28}
+
+[INFO|2025-02-11 01:32:33] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.6241e-07, 'epoch': 2.88, 'throughput': 7895.04}
+
+[INFO|2025-02-11 01:32:57] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.0886e-07, 'epoch': 2.89, 'throughput': 7896.03}
+
+[INFO|2025-02-11 01:33:21] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.5957e-07, 'epoch': 2.90, 'throughput': 7896.57}
+
+[INFO|2025-02-11 01:33:45] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 2.1455e-07, 'epoch': 2.91, 'throughput': 7897.70}
+
+[INFO|2025-02-11 01:34:10] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.7381e-07, 'epoch': 2.92, 'throughput': 7898.05}
+
+[INFO|2025-02-11 01:34:36] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 1.3735e-07, 'epoch': 2.93, 'throughput': 7898.34}
+
+[INFO|2025-02-11 01:35:01] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 1.0517e-07, 'epoch': 2.94, 'throughput': 7898.84}
+
+[INFO|2025-02-11 01:35:26] logging.py:157 >> {'loss': 0.0002, 'learning_rate': 7.7274e-08, 'epoch': 2.94, 'throughput': 7900.67}
+
+[INFO|2025-02-11 01:35:51] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 5.3666e-08, 'epoch': 2.95, 'throughput': 7901.22}
+
+[INFO|2025-02-11 01:36:19] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 3.4349e-08, 'epoch': 2.96, 'throughput': 7899.73}
+
+[INFO|2025-02-11 01:36:44] logging.py:157 >> {'loss': 0.0003, 'learning_rate': 1.9322e-08, 'epoch': 2.97, 'throughput': 7900.03}
+
+[INFO|2025-02-11 01:37:08] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 8.5879e-09, 'epoch': 2.98, 'throughput': 7902.03}
+
+[INFO|2025-02-11 01:37:35] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 2.1470e-09, 'epoch': 2.99, 'throughput': 7903.27}
+
+[INFO|2025-02-11 01:38:00] logging.py:157 >> {'loss': 0.0001, 'learning_rate': 0.0000e+00, 'epoch': 3.00, 'throughput': 7902.86}
+
+[INFO|2025-02-11 01:38:00] trainer.py:3910 >> Saving model checkpoint to saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-339
+
+[INFO|2025-02-11 01:38:00] configuration_utils.py:694 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/config.json
+
+[INFO|2025-02-11 01:38:00] configuration_utils.py:768 >> Model config Qwen2Config {
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-02-11 01:38:00] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-339/tokenizer_config.json
+
+[INFO|2025-02-11 01:38:00] tokenization_utils_base.py:2500 >> Special tokens file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/checkpoint-339/special_tokens_map.json
+
+[INFO|2025-02-11 01:38:01] trainer.py:2643 >> 
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+
+[INFO|2025-02-11 01:38:01] trainer.py:3910 >> Saving model checkpoint to saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128
+
+[INFO|2025-02-11 01:38:01] configuration_utils.py:694 >> loading configuration file /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct/config.json
+
+[INFO|2025-02-11 01:38:01] configuration_utils.py:768 >> Model config Qwen2Config {
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-02-11 01:38:01] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/tokenizer_config.json
+
+[INFO|2025-02-11 01:38:01] tokenization_utils_base.py:2500 >> Special tokens file saved in saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128/special_tokens_map.json
+
+[WARNING|2025-02-11 01:38:02] logging.py:162 >> No metric eval_loss to plot.
+
+[WARNING|2025-02-11 01:38:02] logging.py:162 >> No metric eval_accuracy to plot.
+
+[INFO|2025-02-11 01:38:02] modelcard.py:449 >> Dropping the following result as it does not have all the necessary fields:
+{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
+

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/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": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/tokenizer.json

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

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "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 Qwen, created by Alibaba Cloud. 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 Qwen, created by Alibaba Cloud. You 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": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.997242140099283,
+    "num_input_tokens_seen": 69953200,
+    "total_flos": 2.976146663409713e+18,
+    "train_loss": 0.004280612113766934,
+    "train_runtime": 8852.4475,
+    "train_samples_per_second": 4.914,
+    "train_steps_per_second": 0.038
+}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/trainer_log.jsonl

@@ -0,0 +1,340 @@
+{"current_steps": 1, "total_steps": 339, "loss": 0.2055, "lr": 9.999785297426788e-05, "epoch": 0.00882515168229454, "percentage": 0.29, "elapsed_time": "0:00:27", "remaining_time": "2:35:54", "throughput": 7338.97, "total_tokens": 203120}
+{"current_steps": 2, "total_steps": 339, "loss": 0.1902, "lr": 9.999141208146028e-05, "epoch": 0.01765030336458908, "percentage": 0.59, "elapsed_time": "0:00:54", "remaining_time": "2:32:10", "throughput": 7493.38, "total_tokens": 406048}
+{"current_steps": 3, "total_steps": 339, "loss": 0.1421, "lr": 9.998067787472772e-05, "epoch": 0.026475455046883617, "percentage": 0.88, "elapsed_time": "0:01:18", "remaining_time": "2:26:52", "throughput": 7812.96, "total_tokens": 614736}
+{"current_steps": 4, "total_steps": 339, "loss": 0.1102, "lr": 9.996565127593488e-05, "epoch": 0.03530060672917816, "percentage": 1.18, "elapsed_time": "0:01:46", "remaining_time": "2:28:36", "throughput": 7668.61, "total_tokens": 816416}
+{"current_steps": 5, "total_steps": 339, "loss": 0.0801, "lr": 9.994633357558158e-05, "epoch": 0.0441257584114727, "percentage": 1.47, "elapsed_time": "0:02:12", "remaining_time": "2:27:19", "throughput": 7740.76, "total_tokens": 1024272}
+{"current_steps": 6, "total_steps": 339, "loss": 0.0574, "lr": 9.99227264326918e-05, "epoch": 0.052950910093767234, "percentage": 1.77, "elapsed_time": "0:02:38", "remaining_time": "2:26:42", "throughput": 7743.47, "total_tokens": 1228192}
+{"current_steps": 7, "total_steps": 339, "loss": 0.0401, "lr": 9.989483187467127e-05, "epoch": 0.06177606177606178, "percentage": 2.06, "elapsed_time": "0:03:03", "remaining_time": "2:24:52", "throughput": 7830.06, "total_tokens": 1434992}
+{"current_steps": 8, "total_steps": 339, "loss": 0.0295, "lr": 9.986265229713331e-05, "epoch": 0.07060121345835632, "percentage": 2.36, "elapsed_time": "0:03:31", "remaining_time": "2:25:54", "throughput": 7781.82, "total_tokens": 1646560}
+{"current_steps": 9, "total_steps": 339, "loss": 0.0262, "lr": 9.982619046369321e-05, "epoch": 0.07942636514065085, "percentage": 2.65, "elapsed_time": "0:03:55", "remaining_time": "2:23:41", "throughput": 7819.12, "total_tokens": 1838624}
+{"current_steps": 10, "total_steps": 339, "loss": 0.0263, "lr": 9.978544950573074e-05, "epoch": 0.0882515168229454, "percentage": 2.95, "elapsed_time": "0:04:23", "remaining_time": "2:24:13", "throughput": 7807.27, "total_tokens": 2053488}
+{"current_steps": 11, "total_steps": 339, "loss": 0.022, "lr": 9.974043292212128e-05, "epoch": 0.09707666850523994, "percentage": 3.24, "elapsed_time": "0:04:47", "remaining_time": "2:22:39", "throughput": 7850.91, "total_tokens": 2253680}
+{"current_steps": 12, "total_steps": 339, "loss": 0.0202, "lr": 9.96911445789354e-05, "epoch": 0.10590182018753447, "percentage": 3.54, "elapsed_time": "0:05:11", "remaining_time": "2:21:16", "throughput": 7850.63, "total_tokens": 2442000}
+{"current_steps": 13, "total_steps": 339, "loss": 0.0202, "lr": 9.963758870910671e-05, "epoch": 0.11472697186982901, "percentage": 3.83, "elapsed_time": "0:05:39", "remaining_time": "2:21:53", "throughput": 7823.47, "total_tokens": 2655920}
+{"current_steps": 14, "total_steps": 339, "loss": 0.0178, "lr": 9.957976991206846e-05, "epoch": 0.12355212355212356, "percentage": 4.13, "elapsed_time": "0:06:09", "remaining_time": "2:23:04", "throughput": 7772.01, "total_tokens": 2874064}
+{"current_steps": 15, "total_steps": 339, "loss": 0.0158, "lr": 9.951769315335844e-05, "epoch": 0.13237727523441808, "percentage": 4.42, "elapsed_time": "0:06:33", "remaining_time": "2:21:43", "throughput": 7802.71, "total_tokens": 3071744}
+{"current_steps": 16, "total_steps": 339, "loss": 0.0159, "lr": 9.945136376419259e-05, "epoch": 0.14120242691671264, "percentage": 4.72, "elapsed_time": "0:07:00", "remaining_time": "2:21:38", "throughput": 7786.47, "total_tokens": 3277904}
+{"current_steps": 17, "total_steps": 339, "loss": 0.0147, "lr": 9.938078744100712e-05, "epoch": 0.15002757859900717, "percentage": 5.01, "elapsed_time": "0:07:27", "remaining_time": "2:21:14", "throughput": 7807.8, "total_tokens": 3493136}
+{"current_steps": 18, "total_steps": 339, "loss": 0.0138, "lr": 9.930597024496931e-05, "epoch": 0.1588527302813017, "percentage": 5.31, "elapsed_time": "0:07:54", "remaining_time": "2:20:53", "throughput": 7814.87, "total_tokens": 3704288}
+{"current_steps": 19, "total_steps": 339, "loss": 0.0128, "lr": 9.922691860145696e-05, "epoch": 0.16767788196359626, "percentage": 5.6, "elapsed_time": "0:08:18", "remaining_time": "2:19:51", "throughput": 7835.96, "total_tokens": 3904352}
+{"current_steps": 20, "total_steps": 339, "loss": 0.0124, "lr": 9.914363929950659e-05, "epoch": 0.1765030336458908, "percentage": 5.9, "elapsed_time": "0:08:45", "remaining_time": "2:19:35", "throughput": 7833.94, "total_tokens": 4113888}
+{"current_steps": 21, "total_steps": 339, "loss": 0.0116, "lr": 9.905613949123036e-05, "epoch": 0.18532818532818532, "percentage": 6.19, "elapsed_time": "0:09:11", "remaining_time": "2:19:04", "throughput": 7845.67, "total_tokens": 4323504}
+{"current_steps": 22, "total_steps": 339, "loss": 0.0109, "lr": 9.896442669120187e-05, "epoch": 0.19415333701047988, "percentage": 6.49, "elapsed_time": "0:09:36", "remaining_time": "2:18:27", "throughput": 7844.79, "total_tokens": 4523008}
+{"current_steps": 23, "total_steps": 339, "loss": 0.0106, "lr": 9.886850877581079e-05, "epoch": 0.2029784886927744, "percentage": 6.78, "elapsed_time": "0:10:03", "remaining_time": "2:18:10", "throughput": 7843.06, "total_tokens": 4732864}
+{"current_steps": 24, "total_steps": 339, "loss": 0.0099, "lr": 9.876839398258641e-05, "epoch": 0.21180364037506894, "percentage": 7.08, "elapsed_time": "0:10:30", "remaining_time": "2:17:56", "throughput": 7837.07, "total_tokens": 4941936}
+{"current_steps": 25, "total_steps": 339, "loss": 0.0109, "lr": 9.866409090949022e-05, "epoch": 0.2206287920573635, "percentage": 7.37, "elapsed_time": "0:10:55", "remaining_time": "2:17:11", "throughput": 7848.7, "total_tokens": 5143584}
+{"current_steps": 26, "total_steps": 339, "loss": 0.0084, "lr": 9.855560851417752e-05, "epoch": 0.22945394373965802, "percentage": 7.67, "elapsed_time": "0:11:20", "remaining_time": "2:16:29", "throughput": 7866.16, "total_tokens": 5351024}
+{"current_steps": 27, "total_steps": 339, "loss": 0.0081, "lr": 9.844295611322804e-05, "epoch": 0.23827909542195255, "percentage": 7.96, "elapsed_time": "0:11:46", "remaining_time": "2:16:04", "throughput": 7874.35, "total_tokens": 5563760}
+{"current_steps": 28, "total_steps": 339, "loss": 0.0078, "lr": 9.832614338134595e-05, "epoch": 0.2471042471042471, "percentage": 8.26, "elapsed_time": "0:12:12", "remaining_time": "2:15:39", "throughput": 7877.49, "total_tokens": 5772416}
+{"current_steps": 29, "total_steps": 339, "loss": 0.0081, "lr": 9.820518035052889e-05, "epoch": 0.25592939878654164, "percentage": 8.55, "elapsed_time": "0:12:40", "remaining_time": "2:15:30", "throughput": 7855.15, "total_tokens": 5974464}
+{"current_steps": 30, "total_steps": 339, "loss": 0.0087, "lr": 9.808007740920646e-05, "epoch": 0.26475455046883617, "percentage": 8.85, "elapsed_time": "0:13:07", "remaining_time": "2:15:14", "throughput": 7862.08, "total_tokens": 6193520}
+{"current_steps": 31, "total_steps": 339, "loss": 0.0079, "lr": 9.795084530134801e-05, "epoch": 0.2735797021511307, "percentage": 9.14, "elapsed_time": "0:13:34", "remaining_time": "2:14:50", "throughput": 7859.19, "total_tokens": 6399792}
+{"current_steps": 32, "total_steps": 339, "loss": 0.0086, "lr": 9.781749512553999e-05, "epoch": 0.2824048538334253, "percentage": 9.44, "elapsed_time": "0:14:00", "remaining_time": "2:14:21", "throughput": 7858.66, "total_tokens": 6603584}
+{"current_steps": 33, "total_steps": 339, "loss": 0.0079, "lr": 9.768003833403278e-05, "epoch": 0.2912300055157198, "percentage": 9.73, "elapsed_time": "0:14:27", "remaining_time": "2:14:00", "throughput": 7854.06, "total_tokens": 6810656}
+{"current_steps": 34, "total_steps": 339, "loss": 0.0069, "lr": 9.753848673175707e-05, "epoch": 0.30005515719801434, "percentage": 10.03, "elapsed_time": "0:14:52", "remaining_time": "2:13:26", "throughput": 7844.67, "total_tokens": 7001792}
+{"current_steps": 35, "total_steps": 339, "loss": 0.0064, "lr": 9.739285247531018e-05, "epoch": 0.3088803088803089, "percentage": 10.32, "elapsed_time": "0:15:19", "remaining_time": "2:13:10", "throughput": 7832.92, "total_tokens": 7205952}
+{"current_steps": 36, "total_steps": 339, "loss": 0.006, "lr": 9.724314807191195e-05, "epoch": 0.3177054605626034, "percentage": 10.62, "elapsed_time": "0:15:45", "remaining_time": "2:12:36", "throughput": 7834.75, "total_tokens": 7406304}
+{"current_steps": 37, "total_steps": 339, "loss": 0.0067, "lr": 9.708938637833065e-05, "epoch": 0.32653061224489793, "percentage": 10.91, "elapsed_time": "0:16:12", "remaining_time": "2:12:17", "throughput": 7845.98, "total_tokens": 7629568}
+{"current_steps": 38, "total_steps": 339, "loss": 0.0063, "lr": 9.693158059977878e-05, "epoch": 0.3353557639271925, "percentage": 11.21, "elapsed_time": "0:16:39", "remaining_time": "2:11:53", "throughput": 7852.67, "total_tokens": 7845200}
+{"current_steps": 39, "total_steps": 339, "loss": 0.0058, "lr": 9.676974428877901e-05, "epoch": 0.34418091560948705, "percentage": 11.5, "elapsed_time": "0:17:07", "remaining_time": "2:11:43", "throughput": 7846.29, "total_tokens": 8061840}
+{"current_steps": 40, "total_steps": 339, "loss": 0.0061, "lr": 9.660389134400033e-05, "epoch": 0.3530060672917816, "percentage": 11.8, "elapsed_time": "0:17:36", "remaining_time": "2:11:38", "throughput": 7836.02, "total_tokens": 8279664}
+{"current_steps": 41, "total_steps": 339, "loss": 0.0055, "lr": 9.643403600906433e-05, "epoch": 0.3618312189740761, "percentage": 12.09, "elapsed_time": "0:18:00", "remaining_time": "2:10:56", "throughput": 7840.93, "total_tokens": 8475376}
+{"current_steps": 42, "total_steps": 339, "loss": 0.005, "lr": 9.626019287132203e-05, "epoch": 0.37065637065637064, "percentage": 12.39, "elapsed_time": "0:18:28", "remaining_time": "2:10:37", "throughput": 7842.2, "total_tokens": 8691760}
+{"current_steps": 43, "total_steps": 339, "loss": 0.006, "lr": 9.608237686060099e-05, "epoch": 0.3794815223386652, "percentage": 12.68, "elapsed_time": "0:18:54", "remaining_time": "2:10:07", "throughput": 7833.15, "total_tokens": 8884736}
+{"current_steps": 44, "total_steps": 339, "loss": 0.0048, "lr": 9.590060324792327e-05, "epoch": 0.38830667402095975, "percentage": 12.98, "elapsed_time": "0:19:19", "remaining_time": "2:09:33", "throughput": 7835.19, "total_tokens": 9084064}
+{"current_steps": 45, "total_steps": 339, "loss": 0.0047, "lr": 9.571488764419381e-05, "epoch": 0.3971318257032543, "percentage": 13.27, "elapsed_time": "0:19:46", "remaining_time": "2:09:10", "throughput": 7841.66, "total_tokens": 9302144}
+{"current_steps": 46, "total_steps": 339, "loss": 0.0053, "lr": 9.552524599885981e-05, "epoch": 0.4059569773855488, "percentage": 13.57, "elapsed_time": "0:20:12", "remaining_time": "2:08:44", "throughput": 7848.48, "total_tokens": 9517456}
+{"current_steps": 47, "total_steps": 339, "loss": 0.0044, "lr": 9.533169459854098e-05, "epoch": 0.41478212906784334, "percentage": 13.86, "elapsed_time": "0:20:36", "remaining_time": "2:08:03", "throughput": 7851.65, "total_tokens": 9710768}
+{"current_steps": 48, "total_steps": 339, "loss": 0.0043, "lr": 9.513425006563079e-05, "epoch": 0.42360728075013787, "percentage": 14.16, "elapsed_time": "0:21:02", "remaining_time": "2:07:36", "throughput": 7850.33, "total_tokens": 9914064}
+{"current_steps": 49, "total_steps": 339, "loss": 0.0041, "lr": 9.493292935686895e-05, "epoch": 0.43243243243243246, "percentage": 14.45, "elapsed_time": "0:21:28", "remaining_time": "2:07:06", "throughput": 7853.4, "total_tokens": 10120208}
+{"current_steps": 50, "total_steps": 339, "loss": 0.0044, "lr": 9.472774976188515e-05, "epoch": 0.441257584114727, "percentage": 14.75, "elapsed_time": "0:21:57", "remaining_time": "2:06:57", "throughput": 7850.51, "total_tokens": 10346304}
+{"current_steps": 51, "total_steps": 339, "loss": 0.004, "lr": 9.451872890171419e-05, "epoch": 0.4500827357970215, "percentage": 15.04, "elapsed_time": "0:22:24", "remaining_time": "2:06:30", "throughput": 7847.6, "total_tokens": 10547984}
+{"current_steps": 52, "total_steps": 339, "loss": 0.0045, "lr": 9.43058847272827e-05, "epoch": 0.45890788747931605, "percentage": 15.34, "elapsed_time": "0:22:49", "remaining_time": "2:05:57", "throughput": 7853.35, "total_tokens": 10754288}
+{"current_steps": 53, "total_steps": 339, "loss": 0.0028, "lr": 9.408923551786743e-05, "epoch": 0.4677330391616106, "percentage": 15.63, "elapsed_time": "0:23:14", "remaining_time": "2:05:26", "throughput": 7845.79, "total_tokens": 10942704}
+{"current_steps": 54, "total_steps": 339, "loss": 0.0034, "lr": 9.386879987952549e-05, "epoch": 0.4765581908439051, "percentage": 15.93, "elapsed_time": "0:23:39", "remaining_time": "2:04:53", "throughput": 7853.45, "total_tokens": 11150864}
+{"current_steps": 55, "total_steps": 339, "loss": 0.0042, "lr": 9.364459674349641e-05, "epoch": 0.4853833425261997, "percentage": 16.22, "elapsed_time": "0:24:07", "remaining_time": "2:04:35", "throughput": 7852.19, "total_tokens": 11367728}
+{"current_steps": 56, "total_steps": 339, "loss": 0.0028, "lr": 9.341664536457626e-05, "epoch": 0.4942084942084942, "percentage": 16.52, "elapsed_time": "0:24:31", "remaining_time": "2:03:54", "throughput": 7868.08, "total_tokens": 11575536}
+{"current_steps": 57, "total_steps": 339, "loss": 0.0035, "lr": 9.31849653194641e-05, "epoch": 0.5030336458907887, "percentage": 16.81, "elapsed_time": "0:24:55", "remaining_time": "2:03:21", "throughput": 7875.29, "total_tokens": 11781328}
+{"current_steps": 58, "total_steps": 339, "loss": 0.0029, "lr": 9.294957650508065e-05, "epoch": 0.5118587975730833, "percentage": 17.11, "elapsed_time": "0:25:19", "remaining_time": "2:02:42", "throughput": 7884.45, "total_tokens": 11981232}
+{"current_steps": 59, "total_steps": 339, "loss": 0.0028, "lr": 9.27104991368596e-05, "epoch": 0.5206839492553779, "percentage": 17.4, "elapsed_time": "0:25:44", "remaining_time": "2:02:10", "throughput": 7890.1, "total_tokens": 12187296}
+{"current_steps": 60, "total_steps": 339, "loss": 0.0027, "lr": 9.246775374701139e-05, "epoch": 0.5295091009376723, "percentage": 17.7, "elapsed_time": "0:26:08", "remaining_time": "2:01:34", "throughput": 7895.72, "total_tokens": 12385632}
+{"current_steps": 61, "total_steps": 339, "loss": 0.0022, "lr": 9.222136118275995e-05, "epoch": 0.5383342526199669, "percentage": 17.99, "elapsed_time": "0:26:34", "remaining_time": "2:01:05", "throughput": 7896.89, "total_tokens": 12588928}
+{"current_steps": 62, "total_steps": 339, "loss": 0.0027, "lr": 9.197134260455233e-05, "epoch": 0.5471594043022614, "percentage": 18.29, "elapsed_time": "0:27:05", "remaining_time": "2:01:02", "throughput": 7889.78, "total_tokens": 12825616}
+{"current_steps": 63, "total_steps": 339, "loss": 0.0025, "lr": 9.171771948424137e-05, "epoch": 0.555984555984556, "percentage": 18.58, "elapsed_time": "0:27:33", "remaining_time": "2:00:45", "throughput": 7887.26, "total_tokens": 13044976}
+{"current_steps": 64, "total_steps": 339, "loss": 0.0025, "lr": 9.146051360324166e-05, "epoch": 0.5648097076668506, "percentage": 18.88, "elapsed_time": "0:27:59", "remaining_time": "2:00:14", "throughput": 7894.29, "total_tokens": 13255280}
+{"current_steps": 65, "total_steps": 339, "loss": 0.0022, "lr": 9.119974705065901e-05, "epoch": 0.573634859349145, "percentage": 19.17, "elapsed_time": "0:28:28", "remaining_time": "2:00:03", "throughput": 7878.69, "total_tokens": 13463456}
+{"current_steps": 66, "total_steps": 339, "loss": 0.0023, "lr": 9.093544222139337e-05, "epoch": 0.5824600110314396, "percentage": 19.47, "elapsed_time": "0:28:54", "remaining_time": "1:59:35", "throughput": 7879.18, "total_tokens": 13667744}
+{"current_steps": 67, "total_steps": 339, "loss": 0.0024, "lr": 9.066762181421552e-05, "epoch": 0.5912851627137341, "percentage": 19.76, "elapsed_time": "0:29:21", "remaining_time": "1:59:10", "throughput": 7877.08, "total_tokens": 13874240}
+{"current_steps": 68, "total_steps": 339, "loss": 0.0015, "lr": 9.039630882981768e-05, "epoch": 0.6001103143960287, "percentage": 20.06, "elapsed_time": "0:29:49", "remaining_time": "1:58:53", "throughput": 7867.14, "total_tokens": 14081392}
+{"current_steps": 69, "total_steps": 339, "loss": 0.0033, "lr": 9.012152656883823e-05, "epoch": 0.6089354660783233, "percentage": 20.35, "elapsed_time": "0:30:20", "remaining_time": "1:58:42", "throughput": 7856.5, "total_tokens": 14300896}
+{"current_steps": 70, "total_steps": 339, "loss": 0.0021, "lr": 8.984329862986056e-05, "epoch": 0.6177606177606177, "percentage": 20.65, "elapsed_time": "0:30:47", "remaining_time": "1:58:20", "throughput": 7860.92, "total_tokens": 14523968}
+{"current_steps": 71, "total_steps": 339, "loss": 0.0013, "lr": 8.956164890738643e-05, "epoch": 0.6265857694429123, "percentage": 20.94, "elapsed_time": "0:31:13", "remaining_time": "1:57:50", "throughput": 7863.29, "total_tokens": 14728960}
+{"current_steps": 72, "total_steps": 339, "loss": 0.0016, "lr": 8.927660158978392e-05, "epoch": 0.6354109211252068, "percentage": 21.24, "elapsed_time": "0:31:35", "remaining_time": "1:57:09", "throughput": 7866.76, "total_tokens": 14912480}
+{"current_steps": 73, "total_steps": 339, "loss": 0.0019, "lr": 8.898818115721008e-05, "epoch": 0.6442360728075014, "percentage": 21.53, "elapsed_time": "0:32:00", "remaining_time": "1:56:38", "throughput": 7870.01, "total_tokens": 15114608}
+{"current_steps": 74, "total_steps": 339, "loss": 0.0017, "lr": 8.86964123795085e-05, "epoch": 0.6530612244897959, "percentage": 21.83, "elapsed_time": "0:32:27", "remaining_time": "1:56:13", "throughput": 7870.22, "total_tokens": 15326112}
+{"current_steps": 75, "total_steps": 339, "loss": 0.0015, "lr": 8.84013203140821e-05, "epoch": 0.6618863761720905, "percentage": 22.12, "elapsed_time": "0:32:54", "remaining_time": "1:55:50", "throughput": 7872.22, "total_tokens": 15545248}
+{"current_steps": 76, "total_steps": 339, "loss": 0.0017, "lr": 8.810293030374126e-05, "epoch": 0.670711527854385, "percentage": 22.42, "elapsed_time": "0:33:22", "remaining_time": "1:55:29", "throughput": 7866.0, "total_tokens": 15751872}
+{"current_steps": 77, "total_steps": 339, "loss": 0.001, "lr": 8.780126797452713e-05, "epoch": 0.6795366795366795, "percentage": 22.71, "elapsed_time": "0:33:48", "remaining_time": "1:55:02", "throughput": 7866.6, "total_tokens": 15957872}
+{"current_steps": 78, "total_steps": 339, "loss": 0.0018, "lr": 8.749635923351107e-05, "epoch": 0.6883618312189741, "percentage": 23.01, "elapsed_time": "0:34:13", "remaining_time": "1:54:32", "throughput": 7869.35, "total_tokens": 16162640}
+{"current_steps": 79, "total_steps": 339, "loss": 0.001, "lr": 8.71882302665696e-05, "epoch": 0.6971869829012686, "percentage": 23.3, "elapsed_time": "0:34:37", "remaining_time": "1:53:56", "throughput": 7871.98, "total_tokens": 16352368}
+{"current_steps": 80, "total_steps": 339, "loss": 0.0014, "lr": 8.687690753613554e-05, "epoch": 0.7060121345835632, "percentage": 23.6, "elapsed_time": "0:35:02", "remaining_time": "1:53:27", "throughput": 7877.09, "total_tokens": 16563920}
+{"current_steps": 81, "total_steps": 339, "loss": 0.001, "lr": 8.656241777892543e-05, "epoch": 0.7148372862658577, "percentage": 23.89, "elapsed_time": "0:35:28", "remaining_time": "1:52:58", "throughput": 7874.92, "total_tokens": 16759024}
+{"current_steps": 82, "total_steps": 339, "loss": 0.0013, "lr": 8.624478800364332e-05, "epoch": 0.7236624379481522, "percentage": 24.19, "elapsed_time": "0:35:58", "remaining_time": "1:52:44", "throughput": 7863.86, "total_tokens": 16973728}
+{"current_steps": 83, "total_steps": 339, "loss": 0.0012, "lr": 8.592404548866123e-05, "epoch": 0.7324875896304468, "percentage": 24.48, "elapsed_time": "0:36:22", "remaining_time": "1:52:10", "throughput": 7864.79, "total_tokens": 17162752}
+{"current_steps": 84, "total_steps": 339, "loss": 0.0013, "lr": 8.560021777967649e-05, "epoch": 0.7413127413127413, "percentage": 24.78, "elapsed_time": "0:36:46", "remaining_time": "1:51:37", "throughput": 7871.03, "total_tokens": 17364064}
+{"current_steps": 85, "total_steps": 339, "loss": 0.0011, "lr": 8.527333268734606e-05, "epoch": 0.7501378929950359, "percentage": 25.07, "elapsed_time": "0:37:11", "remaining_time": "1:51:09", "throughput": 7869.86, "total_tokens": 17564576}
+{"current_steps": 86, "total_steps": 339, "loss": 0.0037, "lr": 8.494341828489812e-05, "epoch": 0.7589630446773304, "percentage": 25.37, "elapsed_time": "0:37:40", "remaining_time": "1:50:49", "throughput": 7865.4, "total_tokens": 17778752}
+{"current_steps": 87, "total_steps": 339, "loss": 0.0007, "lr": 8.461050290572114e-05, "epoch": 0.7677881963596249, "percentage": 25.66, "elapsed_time": "0:38:07", "remaining_time": "1:50:26", "throughput": 7860.7, "total_tokens": 17982448}
+{"current_steps": 88, "total_steps": 339, "loss": 0.0008, "lr": 8.427461514093056e-05, "epoch": 0.7766133480419195, "percentage": 25.96, "elapsed_time": "0:38:31", "remaining_time": "1:49:52", "throughput": 7865.68, "total_tokens": 18180608}
+{"current_steps": 89, "total_steps": 339, "loss": 0.0006, "lr": 8.393578383691329e-05, "epoch": 0.785438499724214, "percentage": 26.25, "elapsed_time": "0:38:59", "remaining_time": "1:49:32", "throughput": 7857.75, "total_tokens": 18384496}
+{"current_steps": 90, "total_steps": 339, "loss": 0.001, "lr": 8.359403809285053e-05, "epoch": 0.7942636514065086, "percentage": 26.55, "elapsed_time": "0:39:24", "remaining_time": "1:49:01", "throughput": 7861.64, "total_tokens": 18587744}
+{"current_steps": 91, "total_steps": 339, "loss": 0.001, "lr": 8.324940725821852e-05, "epoch": 0.803088803088803, "percentage": 26.84, "elapsed_time": "0:39:49", "remaining_time": "1:48:31", "throughput": 7864.33, "total_tokens": 18791056}
+{"current_steps": 92, "total_steps": 339, "loss": 0.0008, "lr": 8.290192093026805e-05, "epoch": 0.8119139547710976, "percentage": 27.14, "elapsed_time": "0:40:14", "remaining_time": "1:48:03", "throughput": 7862.08, "total_tokens": 18985008}
+{"current_steps": 93, "total_steps": 339, "loss": 0.0014, "lr": 8.255160895148263e-05, "epoch": 0.8207391064533922, "percentage": 27.43, "elapsed_time": "0:40:39", "remaining_time": "1:47:33", "throughput": 7866.62, "total_tokens": 19193888}
+{"current_steps": 94, "total_steps": 339, "loss": 0.001, "lr": 8.219850140701557e-05, "epoch": 0.8295642581356867, "percentage": 27.73, "elapsed_time": "0:41:05", "remaining_time": "1:47:05", "throughput": 7869.38, "total_tokens": 19399552}
+{"current_steps": 95, "total_steps": 339, "loss": 0.0007, "lr": 8.184262862210624e-05, "epoch": 0.8383894098179813, "percentage": 28.02, "elapsed_time": "0:41:30", "remaining_time": "1:46:36", "throughput": 7872.25, "total_tokens": 19605120}
+{"current_steps": 96, "total_steps": 339, "loss": 0.0008, "lr": 8.148402115947571e-05, "epoch": 0.8472145615002757, "percentage": 28.32, "elapsed_time": "0:41:54", "remaining_time": "1:46:04", "throughput": 7875.39, "total_tokens": 19802480}
+{"current_steps": 97, "total_steps": 339, "loss": 0.0011, "lr": 8.112270981670196e-05, "epoch": 0.8560397131825703, "percentage": 28.61, "elapsed_time": "0:42:22", "remaining_time": "1:45:44", "throughput": 7868.64, "total_tokens": 20009520}
+{"current_steps": 98, "total_steps": 339, "loss": 0.0009, "lr": 8.075872562357501e-05, "epoch": 0.8648648648648649, "percentage": 28.91, "elapsed_time": "0:42:52", "remaining_time": "1:45:26", "throughput": 7865.85, "total_tokens": 20235888}
+{"current_steps": 99, "total_steps": 339, "loss": 0.0006, "lr": 8.039209983943201e-05, "epoch": 0.8736900165471594, "percentage": 29.2, "elapsed_time": "0:43:16", "remaining_time": "1:44:53", "throughput": 7870.61, "total_tokens": 20433600}
+{"current_steps": 100, "total_steps": 339, "loss": 0.0009, "lr": 8.002286395047267e-05, "epoch": 0.882515168229454, "percentage": 29.5, "elapsed_time": "0:43:39", "remaining_time": "1:44:20", "throughput": 7876.56, "total_tokens": 20631664}
+{"current_steps": 101, "total_steps": 339, "loss": 0.0006, "lr": 7.965104966705518e-05, "epoch": 0.8913403199117484, "percentage": 29.79, "elapsed_time": "0:44:04", "remaining_time": "1:43:52", "throughput": 7876.49, "total_tokens": 20833056}
+{"current_steps": 102, "total_steps": 339, "loss": 0.0008, "lr": 7.927668892097289e-05, "epoch": 0.900165471594043, "percentage": 30.09, "elapsed_time": "0:44:33", "remaining_time": "1:43:30", "throughput": 7875.34, "total_tokens": 21051104}
+{"current_steps": 103, "total_steps": 339, "loss": 0.0005, "lr": 7.889981386271201e-05, "epoch": 0.9089906232763376, "percentage": 30.38, "elapsed_time": "0:44:56", "remaining_time": "1:42:59", "throughput": 7878.05, "total_tokens": 21246080}
+{"current_steps": 104, "total_steps": 339, "loss": 0.0006, "lr": 7.852045685869045e-05, "epoch": 0.9178157749586321, "percentage": 30.68, "elapsed_time": "0:45:20", "remaining_time": "1:42:26", "throughput": 7881.27, "total_tokens": 21439696}
+{"current_steps": 105, "total_steps": 339, "loss": 0.0008, "lr": 7.813865048847819e-05, "epoch": 0.9266409266409267, "percentage": 30.97, "elapsed_time": "0:45:48", "remaining_time": "1:42:04", "throughput": 7877.02, "total_tokens": 21648432}
+{"current_steps": 106, "total_steps": 339, "loss": 0.0007, "lr": 7.775442754199928e-05, "epoch": 0.9354660783232212, "percentage": 31.27, "elapsed_time": "0:46:15", "remaining_time": "1:41:41", "throughput": 7877.23, "total_tokens": 21864368}
+{"current_steps": 107, "total_steps": 339, "loss": 0.0006, "lr": 7.736782101671587e-05, "epoch": 0.9442912300055157, "percentage": 31.56, "elapsed_time": "0:46:40", "remaining_time": "1:41:11", "throughput": 7878.63, "total_tokens": 22061968}
+{"current_steps": 108, "total_steps": 339, "loss": 0.0012, "lr": 7.697886411479423e-05, "epoch": 0.9531163816878102, "percentage": 31.86, "elapsed_time": "0:47:06", "remaining_time": "1:40:44", "throughput": 7883.06, "total_tokens": 22278128}
+{"current_steps": 109, "total_steps": 339, "loss": 0.0004, "lr": 7.658759024025349e-05, "epoch": 0.9619415333701048, "percentage": 32.15, "elapsed_time": "0:47:30", "remaining_time": "1:40:15", "throughput": 7881.65, "total_tokens": 22469056}
+{"current_steps": 110, "total_steps": 339, "loss": 0.0008, "lr": 7.619403299609668e-05, "epoch": 0.9707666850523994, "percentage": 32.45, "elapsed_time": "0:47:54", "remaining_time": "1:39:44", "throughput": 7883.6, "total_tokens": 22662128}
+{"current_steps": 111, "total_steps": 339, "loss": 0.0007, "lr": 7.579822618142505e-05, "epoch": 0.9795918367346939, "percentage": 32.74, "elapsed_time": "0:48:22", "remaining_time": "1:39:21", "throughput": 7884.35, "total_tokens": 22883216}
+{"current_steps": 112, "total_steps": 339, "loss": 0.0005, "lr": 7.540020378853523e-05, "epoch": 0.9884169884169884, "percentage": 33.04, "elapsed_time": "0:48:46", "remaining_time": "1:38:50", "throughput": 7889.5, "total_tokens": 23085888}
+{"current_steps": 113, "total_steps": 339, "loss": 0.0007, "lr": 7.500000000000001e-05, "epoch": 0.9972421400992829, "percentage": 33.33, "elapsed_time": "0:49:14", "remaining_time": "1:38:28", "throughput": 7889.54, "total_tokens": 23307520}
+{"current_steps": 114, "total_steps": 339, "loss": 0.0014, "lr": 7.459764918573264e-05, "epoch": 1.0088251516822946, "percentage": 33.63, "elapsed_time": "0:49:46", "remaining_time": "1:38:15", "throughput": 7889.22, "total_tokens": 23564192}
+{"current_steps": 115, "total_steps": 339, "loss": 0.0007, "lr": 7.419318590003523e-05, "epoch": 1.0176503033645892, "percentage": 33.92, "elapsed_time": "0:50:13", "remaining_time": "1:37:50", "throughput": 7886.62, "total_tokens": 23768816}
+{"current_steps": 116, "total_steps": 339, "loss": 0.0006, "lr": 7.378664487863103e-05, "epoch": 1.0264754550468835, "percentage": 34.22, "elapsed_time": "0:50:40", "remaining_time": "1:37:24", "throughput": 7886.1, "total_tokens": 23974096}
+{"current_steps": 117, "total_steps": 339, "loss": 0.0003, "lr": 7.33780610356814e-05, "epoch": 1.0353006067291781, "percentage": 34.51, "elapsed_time": "0:51:05", "remaining_time": "1:36:55", "throughput": 7886.46, "total_tokens": 24172256}
+{"current_steps": 118, "total_steps": 339, "loss": 0.0004, "lr": 7.296746946078736e-05, "epoch": 1.0441257584114727, "percentage": 34.81, "elapsed_time": "0:51:27", "remaining_time": "1:36:23", "throughput": 7889.9, "total_tokens": 24362208}
+{"current_steps": 119, "total_steps": 339, "loss": 0.0003, "lr": 7.255490541597594e-05, "epoch": 1.0529509100937673, "percentage": 35.1, "elapsed_time": "0:51:53", "remaining_time": "1:35:55", "throughput": 7889.45, "total_tokens": 24562224}
+{"current_steps": 120, "total_steps": 339, "loss": 0.0005, "lr": 7.214040433267198e-05, "epoch": 1.0617760617760619, "percentage": 35.4, "elapsed_time": "0:52:21", "remaining_time": "1:35:33", "throughput": 7886.78, "total_tokens": 24776528}
+{"current_steps": 121, "total_steps": 339, "loss": 0.0003, "lr": 7.172400180865513e-05, "epoch": 1.0706012134583562, "percentage": 35.69, "elapsed_time": "0:52:45", "remaining_time": "1:35:02", "throughput": 7893.28, "total_tokens": 24985008}
+{"current_steps": 122, "total_steps": 339, "loss": 0.0005, "lr": 7.130573360500276e-05, "epoch": 1.0794263651406508, "percentage": 35.99, "elapsed_time": "0:53:11", "remaining_time": "1:34:36", "throughput": 7896.08, "total_tokens": 25200720}
+{"current_steps": 123, "total_steps": 339, "loss": 0.0004, "lr": 7.088563564301873e-05, "epoch": 1.0882515168229454, "percentage": 36.28, "elapsed_time": "0:53:36", "remaining_time": "1:34:08", "throughput": 7901.06, "total_tokens": 25413568}
+{"current_steps": 124, "total_steps": 339, "loss": 0.0003, "lr": 7.046374400114842e-05, "epoch": 1.09707666850524, "percentage": 36.58, "elapsed_time": "0:53:59", "remaining_time": "1:33:37", "throughput": 7904.61, "total_tokens": 25608576}
+{"current_steps": 125, "total_steps": 339, "loss": 0.0003, "lr": 7.004009491188022e-05, "epoch": 1.1059018201875346, "percentage": 36.87, "elapsed_time": "0:54:26", "remaining_time": "1:33:11", "throughput": 7904.82, "total_tokens": 25818400}
+{"current_steps": 126, "total_steps": 339, "loss": 0.0005, "lr": 6.961472475863405e-05, "epoch": 1.114726971869829, "percentage": 37.17, "elapsed_time": "0:54:53", "remaining_time": "1:32:47", "throughput": 7905.63, "total_tokens": 26037424}
+{"current_steps": 127, "total_steps": 339, "loss": 0.0005, "lr": 6.918767007263646e-05, "epoch": 1.1235521235521235, "percentage": 37.46, "elapsed_time": "0:55:21", "remaining_time": "1:32:23", "throughput": 7904.23, "total_tokens": 26250480}
+{"current_steps": 128, "total_steps": 339, "loss": 0.0005, "lr": 6.875896752978344e-05, "epoch": 1.1323772752344181, "percentage": 37.76, "elapsed_time": "0:55:47", "remaining_time": "1:31:58", "throughput": 7903.09, "total_tokens": 26458592}
+{"current_steps": 129, "total_steps": 339, "loss": 0.0004, "lr": 6.832865394749065e-05, "epoch": 1.1412024269167127, "percentage": 38.05, "elapsed_time": "0:56:17", "remaining_time": "1:31:38", "throughput": 7898.37, "total_tokens": 26680256}
+{"current_steps": 130, "total_steps": 339, "loss": 0.0004, "lr": 6.789676628153143e-05, "epoch": 1.150027578599007, "percentage": 38.35, "elapsed_time": "0:56:44", "remaining_time": "1:31:13", "throughput": 7897.68, "total_tokens": 26887424}
+{"current_steps": 131, "total_steps": 339, "loss": 0.0003, "lr": 6.746334162286307e-05, "epoch": 1.1588527302813016, "percentage": 38.64, "elapsed_time": "0:57:13", "remaining_time": "1:30:51", "throughput": 7897.31, "total_tokens": 27112736}
+{"current_steps": 132, "total_steps": 339, "loss": 0.0004, "lr": 6.702841719444141e-05, "epoch": 1.1676778819635962, "percentage": 38.94, "elapsed_time": "0:57:40", "remaining_time": "1:30:26", "throughput": 7895.24, "total_tokens": 27320064}
+{"current_steps": 133, "total_steps": 339, "loss": 0.0003, "lr": 6.659203034802397e-05, "epoch": 1.1765030336458908, "percentage": 39.23, "elapsed_time": "0:58:05", "remaining_time": "1:29:58", "throughput": 7895.5, "total_tokens": 27520544}
+{"current_steps": 134, "total_steps": 339, "loss": 0.0009, "lr": 6.615421856096231e-05, "epoch": 1.1853281853281854, "percentage": 39.53, "elapsed_time": "0:58:32", "remaining_time": "1:29:34", "throughput": 7896.09, "total_tokens": 27737920}
+{"current_steps": 135, "total_steps": 339, "loss": 0.0014, "lr": 6.571501943298334e-05, "epoch": 1.19415333701048, "percentage": 39.82, "elapsed_time": "0:59:01", "remaining_time": "1:29:11", "throughput": 7892.01, "total_tokens": 27947552}
+{"current_steps": 136, "total_steps": 339, "loss": 0.0003, "lr": 6.527447068296026e-05, "epoch": 1.2029784886927744, "percentage": 40.12, "elapsed_time": "0:59:26", "remaining_time": "1:28:43", "throughput": 7891.89, "total_tokens": 28143808}
+{"current_steps": 137, "total_steps": 339, "loss": 0.0002, "lr": 6.483261014567311e-05, "epoch": 1.211803640375069, "percentage": 40.41, "elapsed_time": "0:59:53", "remaining_time": "1:28:18", "throughput": 7888.92, "total_tokens": 28349312}
+{"current_steps": 138, "total_steps": 339, "loss": 0.0002, "lr": 6.438947576855968e-05, "epoch": 1.2206287920573635, "percentage": 40.71, "elapsed_time": "1:00:20", "remaining_time": "1:27:53", "throughput": 7888.52, "total_tokens": 28560096}
+{"current_steps": 139, "total_steps": 339, "loss": 0.0005, "lr": 6.394510560845637e-05, "epoch": 1.229453943739658, "percentage": 41.0, "elapsed_time": "1:00:45", "remaining_time": "1:27:25", "throughput": 7890.66, "total_tokens": 28764544}
+{"current_steps": 140, "total_steps": 339, "loss": 0.0004, "lr": 6.349953782832991e-05, "epoch": 1.2382790954219525, "percentage": 41.3, "elapsed_time": "1:01:07", "remaining_time": "1:26:53", "throughput": 7892.64, "total_tokens": 28949360}
+{"current_steps": 141, "total_steps": 339, "loss": 0.0002, "lr": 6.305281069399989e-05, "epoch": 1.247104247104247, "percentage": 41.59, "elapsed_time": "1:01:32", "remaining_time": "1:26:24", "throughput": 7894.92, "total_tokens": 29148112}
+{"current_steps": 142, "total_steps": 339, "loss": 0.0004, "lr": 6.26049625708524e-05, "epoch": 1.2559293987865416, "percentage": 41.89, "elapsed_time": "1:02:01", "remaining_time": "1:26:02", "throughput": 7892.49, "total_tokens": 29370624}
+{"current_steps": 143, "total_steps": 339, "loss": 0.0003, "lr": 6.215603192054522e-05, "epoch": 1.2647545504688362, "percentage": 42.18, "elapsed_time": "1:02:26", "remaining_time": "1:25:34", "throughput": 7894.05, "total_tokens": 29572464}
+{"current_steps": 144, "total_steps": 339, "loss": 0.0006, "lr": 6.17060572977047e-05, "epoch": 1.2735797021511308, "percentage": 42.48, "elapsed_time": "1:02:51", "remaining_time": "1:25:07", "throughput": 7893.7, "total_tokens": 29771152}
+{"current_steps": 145, "total_steps": 339, "loss": 0.0003, "lr": 6.125507734661458e-05, "epoch": 1.2824048538334254, "percentage": 42.77, "elapsed_time": "1:03:13", "remaining_time": "1:24:35", "throughput": 7896.62, "total_tokens": 29954960}
+{"current_steps": 146, "total_steps": 339, "loss": 0.0004, "lr": 6.080313079789723e-05, "epoch": 1.2912300055157198, "percentage": 43.07, "elapsed_time": "1:03:40", "remaining_time": "1:24:10", "throughput": 7895.09, "total_tokens": 30165568}
+{"current_steps": 147, "total_steps": 339, "loss": 0.0005, "lr": 6.035025646518746e-05, "epoch": 1.3000551571980143, "percentage": 43.36, "elapsed_time": "1:04:05", "remaining_time": "1:23:43", "throughput": 7897.58, "total_tokens": 30372160}
+{"current_steps": 148, "total_steps": 339, "loss": 0.0003, "lr": 5.989649324179911e-05, "epoch": 1.308880308880309, "percentage": 43.66, "elapsed_time": "1:04:31", "remaining_time": "1:23:16", "throughput": 7896.39, "total_tokens": 30572752}
+{"current_steps": 149, "total_steps": 339, "loss": 0.0004, "lr": 5.944188009738483e-05, "epoch": 1.3177054605626033, "percentage": 43.95, "elapsed_time": "1:04:57", "remaining_time": "1:22:49", "throughput": 7897.88, "total_tokens": 30780496}
+{"current_steps": 150, "total_steps": 339, "loss": 0.0004, "lr": 5.8986456074589404e-05, "epoch": 1.3265306122448979, "percentage": 44.25, "elapsed_time": "1:05:19", "remaining_time": "1:22:19", "throughput": 7902.0, "total_tokens": 30975120}
+{"current_steps": 151, "total_steps": 339, "loss": 0.0002, "lr": 5.853026028569667e-05, "epoch": 1.3353557639271925, "percentage": 44.54, "elapsed_time": "1:05:45", "remaining_time": "1:21:52", "throughput": 7900.57, "total_tokens": 31174000}
+{"current_steps": 152, "total_steps": 339, "loss": 0.0003, "lr": 5.807333190927053e-05, "epoch": 1.344180915609487, "percentage": 44.84, "elapsed_time": "1:06:12", "remaining_time": "1:21:27", "throughput": 7901.36, "total_tokens": 31387088}
+{"current_steps": 153, "total_steps": 339, "loss": 0.0003, "lr": 5.761571018679025e-05, "epoch": 1.3530060672917816, "percentage": 45.13, "elapsed_time": "1:06:37", "remaining_time": "1:21:00", "throughput": 7898.31, "total_tokens": 31576400}
+{"current_steps": 154, "total_steps": 339, "loss": 0.0003, "lr": 5.715743441928041e-05, "epoch": 1.3618312189740762, "percentage": 45.43, "elapsed_time": "1:07:02", "remaining_time": "1:20:32", "throughput": 7901.98, "total_tokens": 31784320}
+{"current_steps": 155, "total_steps": 339, "loss": 0.0004, "lr": 5.669854396393559e-05, "epoch": 1.3706563706563706, "percentage": 45.72, "elapsed_time": "1:07:29", "remaining_time": "1:20:07", "throughput": 7898.91, "total_tokens": 31987520}
+{"current_steps": 156, "total_steps": 339, "loss": 0.0004, "lr": 5.6239078230740436e-05, "epoch": 1.3794815223386652, "percentage": 46.02, "elapsed_time": "1:07:53", "remaining_time": "1:19:38", "throughput": 7901.45, "total_tokens": 32187456}
+{"current_steps": 157, "total_steps": 339, "loss": 0.0002, "lr": 5.5779076679085054e-05, "epoch": 1.3883066740209598, "percentage": 46.31, "elapsed_time": "1:08:18", "remaining_time": "1:19:11", "throughput": 7900.71, "total_tokens": 32384528}
+{"current_steps": 158, "total_steps": 339, "loss": 0.0004, "lr": 5.531857881437612e-05, "epoch": 1.3971318257032543, "percentage": 46.61, "elapsed_time": "1:08:46", "remaining_time": "1:18:46", "throughput": 7898.98, "total_tokens": 32593040}
+{"current_steps": 159, "total_steps": 339, "loss": 0.0003, "lr": 5.48576241846443e-05, "epoch": 1.4059569773855487, "percentage": 46.9, "elapsed_time": "1:09:12", "remaining_time": "1:18:21", "throughput": 7897.63, "total_tokens": 32797952}
+{"current_steps": 160, "total_steps": 339, "loss": 0.0003, "lr": 5.4396252377147615e-05, "epoch": 1.4147821290678433, "percentage": 47.2, "elapsed_time": "1:09:39", "remaining_time": "1:17:55", "throughput": 7897.71, "total_tokens": 33008800}
+{"current_steps": 161, "total_steps": 339, "loss": 0.0003, "lr": 5.3934503014971793e-05, "epoch": 1.4236072807501379, "percentage": 47.49, "elapsed_time": "1:10:04", "remaining_time": "1:17:28", "throughput": 7897.58, "total_tokens": 33208352}
+{"current_steps": 162, "total_steps": 339, "loss": 0.0002, "lr": 5.347241575362729e-05, "epoch": 1.4324324324324325, "percentage": 47.79, "elapsed_time": "1:10:28", "remaining_time": "1:17:00", "throughput": 7900.98, "total_tokens": 33410208}
+{"current_steps": 163, "total_steps": 339, "loss": 0.0003, "lr": 5.30100302776438e-05, "epoch": 1.441257584114727, "percentage": 48.08, "elapsed_time": "1:10:57", "remaining_time": "1:16:37", "throughput": 7899.49, "total_tokens": 33631888}
+{"current_steps": 164, "total_steps": 339, "loss": 0.0004, "lr": 5.254738629716186e-05, "epoch": 1.4500827357970216, "percentage": 48.38, "elapsed_time": "1:11:21", "remaining_time": "1:16:09", "throughput": 7899.53, "total_tokens": 33825152}
+{"current_steps": 165, "total_steps": 339, "loss": 0.0003, "lr": 5.208452354452274e-05, "epoch": 1.458907887479316, "percentage": 48.67, "elapsed_time": "1:11:45", "remaining_time": "1:15:40", "throughput": 7901.89, "total_tokens": 34020352}
+{"current_steps": 166, "total_steps": 339, "loss": 0.0004, "lr": 5.162148177085604e-05, "epoch": 1.4677330391616106, "percentage": 48.97, "elapsed_time": "1:12:11", "remaining_time": "1:15:13", "throughput": 7902.03, "total_tokens": 34226288}
+{"current_steps": 167, "total_steps": 339, "loss": 0.0016, "lr": 5.115830074266591e-05, "epoch": 1.4765581908439052, "percentage": 49.26, "elapsed_time": "1:12:37", "remaining_time": "1:14:47", "throughput": 7901.05, "total_tokens": 34426672}
+{"current_steps": 168, "total_steps": 339, "loss": 0.0002, "lr": 5.0695020238415756e-05, "epoch": 1.4853833425261997, "percentage": 49.56, "elapsed_time": "1:13:03", "remaining_time": "1:14:21", "throughput": 7901.53, "total_tokens": 34636944}
+{"current_steps": 169, "total_steps": 339, "loss": 0.0003, "lr": 5.0231680045112176e-05, "epoch": 1.494208494208494, "percentage": 49.85, "elapsed_time": "1:13:29", "remaining_time": "1:13:55", "throughput": 7901.79, "total_tokens": 34839456}
+{"current_steps": 170, "total_steps": 339, "loss": 0.0002, "lr": 4.976831995488784e-05, "epoch": 1.5030336458907887, "percentage": 50.15, "elapsed_time": "1:13:53", "remaining_time": "1:13:27", "throughput": 7901.03, "total_tokens": 35031600}
+{"current_steps": 171, "total_steps": 339, "loss": 0.0004, "lr": 4.9304979761584256e-05, "epoch": 1.5118587975730833, "percentage": 50.44, "elapsed_time": "1:14:16", "remaining_time": "1:12:58", "throughput": 7903.96, "total_tokens": 35227728}
+{"current_steps": 172, "total_steps": 339, "loss": 0.0002, "lr": 4.884169925733409e-05, "epoch": 1.5206839492553779, "percentage": 50.74, "elapsed_time": "1:14:42", "remaining_time": "1:12:32", "throughput": 7905.12, "total_tokens": 35436528}
+{"current_steps": 173, "total_steps": 339, "loss": 0.0002, "lr": 4.837851822914397e-05, "epoch": 1.5295091009376725, "percentage": 51.03, "elapsed_time": "1:15:04", "remaining_time": "1:12:02", "throughput": 7909.17, "total_tokens": 35628624}
+{"current_steps": 174, "total_steps": 339, "loss": 0.0002, "lr": 4.791547645547726e-05, "epoch": 1.538334252619967, "percentage": 51.33, "elapsed_time": "1:15:29", "remaining_time": "1:11:35", "throughput": 7909.02, "total_tokens": 35827376}
+{"current_steps": 175, "total_steps": 339, "loss": 0.0003, "lr": 4.745261370283817e-05, "epoch": 1.5471594043022614, "percentage": 51.62, "elapsed_time": "1:16:00", "remaining_time": "1:11:14", "throughput": 7905.59, "total_tokens": 36056560}
+{"current_steps": 176, "total_steps": 339, "loss": 0.0002, "lr": 4.698996972235622e-05, "epoch": 1.555984555984556, "percentage": 51.92, "elapsed_time": "1:16:27", "remaining_time": "1:10:48", "throughput": 7905.47, "total_tokens": 36267568}
+{"current_steps": 177, "total_steps": 339, "loss": 0.0027, "lr": 4.652758424637271e-05, "epoch": 1.5648097076668506, "percentage": 52.21, "elapsed_time": "1:16:54", "remaining_time": "1:10:23", "throughput": 7904.7, "total_tokens": 36473008}
+{"current_steps": 178, "total_steps": 339, "loss": 0.0004, "lr": 4.606549698502823e-05, "epoch": 1.573634859349145, "percentage": 52.51, "elapsed_time": "1:17:18", "remaining_time": "1:09:55", "throughput": 7905.8, "total_tokens": 36670944}
+{"current_steps": 179, "total_steps": 339, "loss": 0.0001, "lr": 4.56037476228524e-05, "epoch": 1.5824600110314395, "percentage": 52.8, "elapsed_time": "1:17:44", "remaining_time": "1:09:29", "throughput": 7907.65, "total_tokens": 36882256}
+{"current_steps": 180, "total_steps": 339, "loss": 0.0004, "lr": 4.5142375815355706e-05, "epoch": 1.591285162713734, "percentage": 53.1, "elapsed_time": "1:18:10", "remaining_time": "1:09:03", "throughput": 7907.43, "total_tokens": 37091392}
+{"current_steps": 181, "total_steps": 339, "loss": 0.0002, "lr": 4.468142118562389e-05, "epoch": 1.6001103143960287, "percentage": 53.39, "elapsed_time": "1:18:41", "remaining_time": "1:08:41", "throughput": 7902.86, "total_tokens": 37309680}
+{"current_steps": 182, "total_steps": 339, "loss": 0.0003, "lr": 4.4220923320914964e-05, "epoch": 1.6089354660783233, "percentage": 53.69, "elapsed_time": "1:19:07", "remaining_time": "1:08:15", "throughput": 7903.1, "total_tokens": 37517952}
+{"current_steps": 183, "total_steps": 339, "loss": 0.0003, "lr": 4.376092176925958e-05, "epoch": 1.6177606177606179, "percentage": 53.98, "elapsed_time": "1:19:33", "remaining_time": "1:07:49", "throughput": 7904.08, "total_tokens": 37732160}
+{"current_steps": 184, "total_steps": 339, "loss": 0.0004, "lr": 4.330145603606441e-05, "epoch": 1.6265857694429124, "percentage": 54.28, "elapsed_time": "1:19:59", "remaining_time": "1:07:22", "throughput": 7905.25, "total_tokens": 37940368}
+{"current_steps": 185, "total_steps": 339, "loss": 0.0004, "lr": 4.2842565580719595e-05, "epoch": 1.6354109211252068, "percentage": 54.57, "elapsed_time": "1:20:21", "remaining_time": "1:06:53", "throughput": 7908.78, "total_tokens": 38135024}
+{"current_steps": 186, "total_steps": 339, "loss": 0.0002, "lr": 4.238428981320975e-05, "epoch": 1.6442360728075014, "percentage": 54.87, "elapsed_time": "1:20:45", "remaining_time": "1:06:25", "throughput": 7912.45, "total_tokens": 38336176}
+{"current_steps": 187, "total_steps": 339, "loss": 0.0003, "lr": 4.192666809072948e-05, "epoch": 1.6530612244897958, "percentage": 55.16, "elapsed_time": "1:21:11", "remaining_time": "1:05:59", "throughput": 7912.95, "total_tokens": 38548880}
+{"current_steps": 188, "total_steps": 339, "loss": 0.0003, "lr": 4.146973971430333e-05, "epoch": 1.6618863761720903, "percentage": 55.46, "elapsed_time": "1:21:37", "remaining_time": "1:05:33", "throughput": 7913.14, "total_tokens": 38755920}
+{"current_steps": 189, "total_steps": 339, "loss": 0.0002, "lr": 4.101354392541061e-05, "epoch": 1.670711527854385, "percentage": 55.75, "elapsed_time": "1:22:04", "remaining_time": "1:05:08", "throughput": 7914.08, "total_tokens": 38973328}
+{"current_steps": 190, "total_steps": 339, "loss": 0.0003, "lr": 4.0558119902615174e-05, "epoch": 1.6795366795366795, "percentage": 56.05, "elapsed_time": "1:22:34", "remaining_time": "1:04:45", "throughput": 7911.16, "total_tokens": 39193232}
+{"current_steps": 191, "total_steps": 339, "loss": 0.0003, "lr": 4.010350675820091e-05, "epoch": 1.688361831218974, "percentage": 56.34, "elapsed_time": "1:23:00", "remaining_time": "1:04:19", "throughput": 7911.74, "total_tokens": 39406608}
+{"current_steps": 192, "total_steps": 339, "loss": 0.0004, "lr": 3.964974353481254e-05, "epoch": 1.6971869829012687, "percentage": 56.64, "elapsed_time": "1:23:28", "remaining_time": "1:03:54", "throughput": 7911.05, "total_tokens": 39620160}
+{"current_steps": 193, "total_steps": 339, "loss": 0.0001, "lr": 3.919686920210277e-05, "epoch": 1.7060121345835633, "percentage": 56.93, "elapsed_time": "1:23:53", "remaining_time": "1:03:27", "throughput": 7910.45, "total_tokens": 39815952}
+{"current_steps": 194, "total_steps": 339, "loss": 0.0003, "lr": 3.874492265338544e-05, "epoch": 1.7148372862658579, "percentage": 57.23, "elapsed_time": "1:24:17", "remaining_time": "1:03:00", "throughput": 7911.47, "total_tokens": 40015408}
+{"current_steps": 195, "total_steps": 339, "loss": 0.0002, "lr": 3.829394270229531e-05, "epoch": 1.7236624379481522, "percentage": 57.52, "elapsed_time": "1:24:42", "remaining_time": "1:02:32", "throughput": 7913.07, "total_tokens": 40215328}
+{"current_steps": 196, "total_steps": 339, "loss": 0.0002, "lr": 3.784396807945477e-05, "epoch": 1.7324875896304468, "percentage": 57.82, "elapsed_time": "1:25:07", "remaining_time": "1:02:06", "throughput": 7912.61, "total_tokens": 40414384}
+{"current_steps": 197, "total_steps": 339, "loss": 0.0002, "lr": 3.7395037429147615e-05, "epoch": 1.7413127413127412, "percentage": 58.11, "elapsed_time": "1:25:32", "remaining_time": "1:01:39", "throughput": 7914.15, "total_tokens": 40620656}
+{"current_steps": 198, "total_steps": 339, "loss": 0.0003, "lr": 3.694718930600012e-05, "epoch": 1.7501378929950357, "percentage": 58.41, "elapsed_time": "1:26:04", "remaining_time": "1:01:17", "throughput": 7908.9, "total_tokens": 40847008}
+{"current_steps": 199, "total_steps": 339, "loss": 0.0001, "lr": 3.65004621716701e-05, "epoch": 1.7589630446773303, "percentage": 58.7, "elapsed_time": "1:26:27", "remaining_time": "1:00:49", "throughput": 7909.88, "total_tokens": 41036368}
+{"current_steps": 200, "total_steps": 339, "loss": 0.0003, "lr": 3.6054894391543646e-05, "epoch": 1.767788196359625, "percentage": 59.0, "elapsed_time": "1:26:59", "remaining_time": "1:00:27", "throughput": 7904.23, "total_tokens": 41252976}
+{"current_steps": 201, "total_steps": 339, "loss": 0.0002, "lr": 3.561052423144032e-05, "epoch": 1.7766133480419195, "percentage": 59.29, "elapsed_time": "1:27:29", "remaining_time": "1:00:04", "throughput": 7898.91, "total_tokens": 41465104}
+{"current_steps": 202, "total_steps": 339, "loss": 0.0002, "lr": 3.5167389854326905e-05, "epoch": 1.785438499724214, "percentage": 59.59, "elapsed_time": "1:27:56", "remaining_time": "0:59:38", "throughput": 7897.44, "total_tokens": 41670800}
+{"current_steps": 203, "total_steps": 339, "loss": 0.0013, "lr": 3.4725529317039754e-05, "epoch": 1.7942636514065087, "percentage": 59.88, "elapsed_time": "1:28:25", "remaining_time": "0:59:14", "throughput": 7894.63, "total_tokens": 41883536}
+{"current_steps": 204, "total_steps": 339, "loss": 0.0001, "lr": 3.428498056701665e-05, "epoch": 1.803088803088803, "percentage": 60.18, "elapsed_time": "1:28:51", "remaining_time": "0:58:48", "throughput": 7893.32, "total_tokens": 42083360}
+{"current_steps": 205, "total_steps": 339, "loss": 0.0002, "lr": 3.38457814390377e-05, "epoch": 1.8119139547710976, "percentage": 60.47, "elapsed_time": "1:29:16", "remaining_time": "0:58:21", "throughput": 7893.48, "total_tokens": 42283120}
+{"current_steps": 206, "total_steps": 339, "loss": 0.0003, "lr": 3.340796965197604e-05, "epoch": 1.8207391064533922, "percentage": 60.77, "elapsed_time": "1:29:41", "remaining_time": "0:57:54", "throughput": 7897.32, "total_tokens": 42499088}
+{"current_steps": 207, "total_steps": 339, "loss": 0.0001, "lr": 3.297158280555862e-05, "epoch": 1.8295642581356866, "percentage": 61.06, "elapsed_time": "1:30:05", "remaining_time": "0:57:26", "throughput": 7898.07, "total_tokens": 42692976}
+{"current_steps": 208, "total_steps": 339, "loss": 0.0003, "lr": 3.2536658377136935e-05, "epoch": 1.8383894098179812, "percentage": 61.36, "elapsed_time": "1:30:32", "remaining_time": "0:57:01", "throughput": 7898.41, "total_tokens": 42907216}
+{"current_steps": 209, "total_steps": 339, "loss": 0.0001, "lr": 3.210323371846857e-05, "epoch": 1.8472145615002757, "percentage": 61.65, "elapsed_time": "1:30:58", "remaining_time": "0:56:35", "throughput": 7898.41, "total_tokens": 43112448}
+{"current_steps": 210, "total_steps": 339, "loss": 0.0003, "lr": 3.167134605250938e-05, "epoch": 1.8560397131825703, "percentage": 61.95, "elapsed_time": "1:31:27", "remaining_time": "0:56:10", "throughput": 7897.79, "total_tokens": 43340096}
+{"current_steps": 211, "total_steps": 339, "loss": 0.0001, "lr": 3.124103247021657e-05, "epoch": 1.864864864864865, "percentage": 62.24, "elapsed_time": "1:31:53", "remaining_time": "0:55:44", "throughput": 7896.37, "total_tokens": 43539664}
+{"current_steps": 212, "total_steps": 339, "loss": 0.0003, "lr": 3.081232992736355e-05, "epoch": 1.8736900165471595, "percentage": 62.54, "elapsed_time": "1:32:16", "remaining_time": "0:55:16", "throughput": 7897.48, "total_tokens": 43727664}
+{"current_steps": 213, "total_steps": 339, "loss": 0.0002, "lr": 3.0385275241365962e-05, "epoch": 1.882515168229454, "percentage": 62.83, "elapsed_time": "1:32:44", "remaining_time": "0:54:51", "throughput": 7898.39, "total_tokens": 43953584}
+{"current_steps": 214, "total_steps": 339, "loss": 0.0002, "lr": 2.9959905088119776e-05, "epoch": 1.8913403199117484, "percentage": 63.13, "elapsed_time": "1:33:11", "remaining_time": "0:54:25", "throughput": 7897.6, "total_tokens": 44157504}
+{"current_steps": 215, "total_steps": 339, "loss": 0.0001, "lr": 2.9536255998851613e-05, "epoch": 1.900165471594043, "percentage": 63.42, "elapsed_time": "1:33:35", "remaining_time": "0:53:58", "throughput": 7898.54, "total_tokens": 44350448}
+{"current_steps": 216, "total_steps": 339, "loss": 0.0002, "lr": 2.9114364356981272e-05, "epoch": 1.9089906232763376, "percentage": 63.72, "elapsed_time": "1:33:59", "remaining_time": "0:53:31", "throughput": 7900.99, "total_tokens": 44561472}
+{"current_steps": 217, "total_steps": 339, "loss": 0.0002, "lr": 2.8694266394997238e-05, "epoch": 1.917815774958632, "percentage": 64.01, "elapsed_time": "1:34:27", "remaining_time": "0:53:06", "throughput": 7899.67, "total_tokens": 44769936}
+{"current_steps": 218, "total_steps": 339, "loss": 0.0002, "lr": 2.8275998191344888e-05, "epoch": 1.9266409266409266, "percentage": 64.31, "elapsed_time": "1:34:54", "remaining_time": "0:52:40", "throughput": 7898.82, "total_tokens": 44979344}
+{"current_steps": 219, "total_steps": 339, "loss": 0.0002, "lr": 2.7859595667328026e-05, "epoch": 1.9354660783232212, "percentage": 64.6, "elapsed_time": "1:35:22", "remaining_time": "0:52:15", "throughput": 7898.4, "total_tokens": 45196944}
+{"current_steps": 220, "total_steps": 339, "loss": 0.0001, "lr": 2.7445094584024067e-05, "epoch": 1.9442912300055157, "percentage": 64.9, "elapsed_time": "1:35:51", "remaining_time": "0:51:50", "throughput": 7895.44, "total_tokens": 45406832}
+{"current_steps": 221, "total_steps": 339, "loss": 0.0003, "lr": 2.7032530539212658e-05, "epoch": 1.9531163816878103, "percentage": 65.19, "elapsed_time": "1:36:18", "remaining_time": "0:51:25", "throughput": 7892.18, "total_tokens": 45603120}
+{"current_steps": 222, "total_steps": 339, "loss": 0.0002, "lr": 2.6621938964318595e-05, "epoch": 1.961941533370105, "percentage": 65.49, "elapsed_time": "1:36:44", "remaining_time": "0:50:59", "throughput": 7890.65, "total_tokens": 45805184}
+{"current_steps": 223, "total_steps": 339, "loss": 0.0001, "lr": 2.621335512136899e-05, "epoch": 1.9707666850523995, "percentage": 65.78, "elapsed_time": "1:37:08", "remaining_time": "0:50:31", "throughput": 7892.67, "total_tokens": 46001184}
+{"current_steps": 224, "total_steps": 339, "loss": 0.0002, "lr": 2.5806814099964772e-05, "epoch": 1.9795918367346939, "percentage": 66.08, "elapsed_time": "1:37:34", "remaining_time": "0:50:05", "throughput": 7892.1, "total_tokens": 46206288}
+{"current_steps": 225, "total_steps": 339, "loss": 0.0002, "lr": 2.540235081426736e-05, "epoch": 1.9884169884169884, "percentage": 66.37, "elapsed_time": "1:38:03", "remaining_time": "0:49:40", "throughput": 7891.47, "total_tokens": 46427344}
+{"current_steps": 226, "total_steps": 339, "loss": 0.0003, "lr": 2.500000000000001e-05, "epoch": 1.9972421400992828, "percentage": 66.67, "elapsed_time": "1:38:28", "remaining_time": "0:49:14", "throughput": 7891.65, "total_tokens": 46627344}
+{"current_steps": 227, "total_steps": 339, "loss": 0.0011, "lr": 2.459979621146477e-05, "epoch": 2.0088251516822946, "percentage": 66.96, "elapsed_time": "1:39:03", "remaining_time": "0:48:52", "throughput": 7890.98, "total_tokens": 46901504}
+{"current_steps": 228, "total_steps": 339, "loss": 0.0001, "lr": 2.4201773818574956e-05, "epoch": 2.017650303364589, "percentage": 67.26, "elapsed_time": "1:39:31", "remaining_time": "0:48:27", "throughput": 7888.59, "total_tokens": 47104400}
+{"current_steps": 229, "total_steps": 339, "loss": 0.0001, "lr": 2.3805967003903333e-05, "epoch": 2.0264754550468838, "percentage": 67.55, "elapsed_time": "1:39:56", "remaining_time": "0:48:00", "throughput": 7889.79, "total_tokens": 47314176}
+{"current_steps": 230, "total_steps": 339, "loss": 0.0003, "lr": 2.3412409759746528e-05, "epoch": 2.0353006067291783, "percentage": 67.85, "elapsed_time": "1:40:25", "remaining_time": "0:47:35", "throughput": 7887.38, "total_tokens": 47525264}
+{"current_steps": 231, "total_steps": 339, "loss": 0.0001, "lr": 2.302113588520578e-05, "epoch": 2.0441257584114725, "percentage": 68.14, "elapsed_time": "1:40:49", "remaining_time": "0:47:08", "throughput": 7889.45, "total_tokens": 47724528}
+{"current_steps": 232, "total_steps": 339, "loss": 0.0002, "lr": 2.2632178983284153e-05, "epoch": 2.052950910093767, "percentage": 68.44, "elapsed_time": "1:41:17", "remaining_time": "0:46:42", "throughput": 7887.07, "total_tokens": 47932624}
+{"current_steps": 233, "total_steps": 339, "loss": 0.0001, "lr": 2.2245572458000712e-05, "epoch": 2.0617760617760617, "percentage": 68.73, "elapsed_time": "1:41:42", "remaining_time": "0:46:16", "throughput": 7889.59, "total_tokens": 48148608}
+{"current_steps": 234, "total_steps": 339, "loss": 0.0025, "lr": 2.1861349511521815e-05, "epoch": 2.0706012134583562, "percentage": 69.03, "elapsed_time": "1:42:14", "remaining_time": "0:45:52", "throughput": 7885.55, "total_tokens": 48373632}
+{"current_steps": 235, "total_steps": 339, "loss": 0.0013, "lr": 2.147954314130955e-05, "epoch": 2.079426365140651, "percentage": 69.32, "elapsed_time": "1:42:44", "remaining_time": "0:45:28", "throughput": 7881.47, "total_tokens": 48586512}
+{"current_steps": 236, "total_steps": 339, "loss": 0.0001, "lr": 2.1100186137288e-05, "epoch": 2.0882515168229454, "percentage": 69.62, "elapsed_time": "1:43:08", "remaining_time": "0:45:01", "throughput": 7884.07, "total_tokens": 48793568}
+{"current_steps": 237, "total_steps": 339, "loss": 0.0001, "lr": 2.072331107902713e-05, "epoch": 2.09707666850524, "percentage": 69.91, "elapsed_time": "1:43:35", "remaining_time": "0:44:34", "throughput": 7884.69, "total_tokens": 49006224}
+{"current_steps": 238, "total_steps": 339, "loss": 0.0002, "lr": 2.0348950332944834e-05, "epoch": 2.1059018201875346, "percentage": 70.21, "elapsed_time": "1:44:00", "remaining_time": "0:44:08", "throughput": 7886.69, "total_tokens": 49217632}
+{"current_steps": 239, "total_steps": 339, "loss": 0.0001, "lr": 1.9977136049527345e-05, "epoch": 2.114726971869829, "percentage": 70.5, "elapsed_time": "1:44:26", "remaining_time": "0:43:42", "throughput": 7886.9, "total_tokens": 49426624}
+{"current_steps": 240, "total_steps": 339, "loss": 0.0001, "lr": 1.960790016056801e-05, "epoch": 2.1235521235521237, "percentage": 70.8, "elapsed_time": "1:44:50", "remaining_time": "0:43:14", "throughput": 7888.14, "total_tokens": 49623376}
+{"current_steps": 241, "total_steps": 339, "loss": 0.0002, "lr": 1.9241274376425e-05, "epoch": 2.132377275234418, "percentage": 71.09, "elapsed_time": "1:45:16", "remaining_time": "0:42:48", "throughput": 7889.13, "total_tokens": 49828144}
+{"current_steps": 242, "total_steps": 339, "loss": 0.0002, "lr": 1.8877290183298057e-05, "epoch": 2.1412024269167125, "percentage": 71.39, "elapsed_time": "1:45:40", "remaining_time": "0:42:21", "throughput": 7888.3, "total_tokens": 50018448}
+{"current_steps": 243, "total_steps": 339, "loss": 0.0001, "lr": 1.8515978840524302e-05, "epoch": 2.150027578599007, "percentage": 71.68, "elapsed_time": "1:46:06", "remaining_time": "0:41:55", "throughput": 7887.77, "total_tokens": 50218176}
+{"current_steps": 244, "total_steps": 339, "loss": 0.0002, "lr": 1.815737137789377e-05, "epoch": 2.1588527302813016, "percentage": 71.98, "elapsed_time": "1:46:31", "remaining_time": "0:41:28", "throughput": 7889.8, "total_tokens": 50424896}
+{"current_steps": 245, "total_steps": 339, "loss": 0.0006, "lr": 1.7801498592984446e-05, "epoch": 2.1676778819635962, "percentage": 72.27, "elapsed_time": "1:46:59", "remaining_time": "0:41:03", "throughput": 7887.66, "total_tokens": 50635088}
+{"current_steps": 246, "total_steps": 339, "loss": 0.0001, "lr": 1.7448391048517376e-05, "epoch": 2.176503033645891, "percentage": 72.57, "elapsed_time": "1:47:27", "remaining_time": "0:40:37", "throughput": 7886.91, "total_tokens": 50849552}
+{"current_steps": 247, "total_steps": 339, "loss": 0.0002, "lr": 1.7098079069731958e-05, "epoch": 2.1853281853281854, "percentage": 72.86, "elapsed_time": "1:47:49", "remaining_time": "0:40:09", "throughput": 7889.08, "total_tokens": 51037776}
+{"current_steps": 248, "total_steps": 339, "loss": 0.0002, "lr": 1.6750592741781497e-05, "epoch": 2.19415333701048, "percentage": 73.16, "elapsed_time": "1:48:14", "remaining_time": "0:39:42", "throughput": 7890.61, "total_tokens": 51242672}
+{"current_steps": 249, "total_steps": 339, "loss": 0.0001, "lr": 1.640596190714947e-05, "epoch": 2.2029784886927746, "percentage": 73.45, "elapsed_time": "1:48:39", "remaining_time": "0:39:16", "throughput": 7889.22, "total_tokens": 51437008}
+{"current_steps": 250, "total_steps": 339, "loss": 0.0001, "lr": 1.6064216163086716e-05, "epoch": 2.211803640375069, "percentage": 73.75, "elapsed_time": "1:49:04", "remaining_time": "0:38:49", "throughput": 7890.88, "total_tokens": 51641264}
+{"current_steps": 251, "total_steps": 339, "loss": 0.0001, "lr": 1.5725384859069455e-05, "epoch": 2.2206287920573633, "percentage": 74.04, "elapsed_time": "1:49:30", "remaining_time": "0:38:23", "throughput": 7890.72, "total_tokens": 51842592}
+{"current_steps": 252, "total_steps": 339, "loss": 0.0012, "lr": 1.538949709427886e-05, "epoch": 2.229453943739658, "percentage": 74.34, "elapsed_time": "1:49:55", "remaining_time": "0:37:57", "throughput": 7890.97, "total_tokens": 52047456}
+{"current_steps": 253, "total_steps": 339, "loss": 0.0001, "lr": 1.5056581715101886e-05, "epoch": 2.2382790954219525, "percentage": 74.63, "elapsed_time": "1:50:19", "remaining_time": "0:37:30", "throughput": 7892.37, "total_tokens": 52242208}
+{"current_steps": 254, "total_steps": 339, "loss": 0.0003, "lr": 1.472666731265394e-05, "epoch": 2.247104247104247, "percentage": 74.93, "elapsed_time": "1:50:42", "remaining_time": "0:37:02", "throughput": 7894.16, "total_tokens": 52436800}
+{"current_steps": 255, "total_steps": 339, "loss": 0.0001, "lr": 1.4399782220323515e-05, "epoch": 2.2559293987865416, "percentage": 75.22, "elapsed_time": "1:51:06", "remaining_time": "0:36:35", "throughput": 7894.39, "total_tokens": 52624752}
+{"current_steps": 256, "total_steps": 339, "loss": 0.0001, "lr": 1.4075954511338785e-05, "epoch": 2.2647545504688362, "percentage": 75.52, "elapsed_time": "1:51:32", "remaining_time": "0:36:09", "throughput": 7895.04, "total_tokens": 52836384}
+{"current_steps": 257, "total_steps": 339, "loss": 0.0001, "lr": 1.3755211996356687e-05, "epoch": 2.273579702151131, "percentage": 75.81, "elapsed_time": "1:52:00", "remaining_time": "0:35:44", "throughput": 7894.62, "total_tokens": 53059296}
+{"current_steps": 258, "total_steps": 339, "loss": 0.0001, "lr": 1.3437582221074573e-05, "epoch": 2.2824048538334254, "percentage": 76.11, "elapsed_time": "1:52:28", "remaining_time": "0:35:18", "throughput": 7893.02, "total_tokens": 53267440}
+{"current_steps": 259, "total_steps": 339, "loss": 0.0001, "lr": 1.3123092463864456e-05, "epoch": 2.29123000551572, "percentage": 76.4, "elapsed_time": "1:52:59", "remaining_time": "0:34:54", "throughput": 7891.68, "total_tokens": 53501008}
+{"current_steps": 260, "total_steps": 339, "loss": 0.0001, "lr": 1.2811769733430406e-05, "epoch": 2.300055157198014, "percentage": 76.7, "elapsed_time": "1:53:24", "remaining_time": "0:34:27", "throughput": 7892.09, "total_tokens": 53700432}
+{"current_steps": 261, "total_steps": 339, "loss": 0.0002, "lr": 1.250364076648894e-05, "epoch": 2.3088803088803087, "percentage": 76.99, "elapsed_time": "1:53:50", "remaining_time": "0:34:01", "throughput": 7894.29, "total_tokens": 53919616}
+{"current_steps": 262, "total_steps": 339, "loss": 0.0001, "lr": 1.2198732025472876e-05, "epoch": 2.3177054605626033, "percentage": 77.29, "elapsed_time": "1:54:17", "remaining_time": "0:33:35", "throughput": 7893.79, "total_tokens": 54130528}
+{"current_steps": 263, "total_steps": 339, "loss": 0.0002, "lr": 1.1897069696258755e-05, "epoch": 2.326530612244898, "percentage": 77.58, "elapsed_time": "1:54:44", "remaining_time": "0:33:09", "throughput": 7894.54, "total_tokens": 54350560}
+{"current_steps": 264, "total_steps": 339, "loss": 0.0001, "lr": 1.1598679685917901e-05, "epoch": 2.3353557639271925, "percentage": 77.88, "elapsed_time": "1:55:08", "remaining_time": "0:32:42", "throughput": 7894.72, "total_tokens": 54542224}
+{"current_steps": 265, "total_steps": 339, "loss": 0.0002, "lr": 1.1303587620491513e-05, "epoch": 2.344180915609487, "percentage": 78.17, "elapsed_time": "1:55:35", "remaining_time": "0:32:16", "throughput": 7893.77, "total_tokens": 54745136}
+{"current_steps": 266, "total_steps": 339, "loss": 0.0001, "lr": 1.1011818842789928e-05, "epoch": 2.3530060672917816, "percentage": 78.47, "elapsed_time": "1:56:01", "remaining_time": "0:31:50", "throughput": 7895.0, "total_tokens": 54957584}
+{"current_steps": 267, "total_steps": 339, "loss": 0.0001, "lr": 1.0723398410216084e-05, "epoch": 2.361831218974076, "percentage": 78.76, "elapsed_time": "1:56:26", "remaining_time": "0:31:24", "throughput": 7895.05, "total_tokens": 55162496}
+{"current_steps": 268, "total_steps": 339, "loss": 0.0002, "lr": 1.0438351092613569e-05, "epoch": 2.370656370656371, "percentage": 79.06, "elapsed_time": "1:56:53", "remaining_time": "0:30:58", "throughput": 7895.15, "total_tokens": 55376544}
+{"current_steps": 269, "total_steps": 339, "loss": 0.0001, "lr": 1.0156701370139454e-05, "epoch": 2.3794815223386654, "percentage": 79.35, "elapsed_time": "1:57:19", "remaining_time": "0:30:31", "throughput": 7895.9, "total_tokens": 55583072}
+{"current_steps": 270, "total_steps": 339, "loss": 0.0002, "lr": 9.878473431161767e-06, "epoch": 2.38830667402096, "percentage": 79.65, "elapsed_time": "1:57:46", "remaining_time": "0:30:05", "throughput": 7897.0, "total_tokens": 55801200}
+{"current_steps": 271, "total_steps": 339, "loss": 0.0001, "lr": 9.603691170182317e-06, "epoch": 2.397131825703254, "percentage": 79.94, "elapsed_time": "1:58:14", "remaining_time": "0:29:40", "throughput": 7893.16, "total_tokens": 55998080}
+{"current_steps": 272, "total_steps": 339, "loss": 0.0002, "lr": 9.33237818578449e-06, "epoch": 2.4059569773855487, "percentage": 80.24, "elapsed_time": "1:58:41", "remaining_time": "0:29:14", "throughput": 7891.24, "total_tokens": 56200448}
+{"current_steps": 273, "total_steps": 339, "loss": 0.0001, "lr": 9.064557778606631e-06, "epoch": 2.4147821290678433, "percentage": 80.53, "elapsed_time": "1:59:05", "remaining_time": "0:28:47", "throughput": 7893.3, "total_tokens": 56400416}
+{"current_steps": 274, "total_steps": 339, "loss": 0.0002, "lr": 8.800252949340998e-06, "epoch": 2.423607280750138, "percentage": 80.83, "elapsed_time": "1:59:31", "remaining_time": "0:28:21", "throughput": 7893.02, "total_tokens": 56606128}
+{"current_steps": 275, "total_steps": 339, "loss": 0.0, "lr": 8.539486396758356e-06, "epoch": 2.4324324324324325, "percentage": 81.12, "elapsed_time": "1:59:56", "remaining_time": "0:27:54", "throughput": 7892.52, "total_tokens": 56797824}
+{"current_steps": 276, "total_steps": 339, "loss": 0.0001, "lr": 8.28228051575864e-06, "epoch": 2.441257584114727, "percentage": 81.42, "elapsed_time": "2:00:22", "remaining_time": "0:27:28", "throughput": 7892.63, "total_tokens": 57006384}
+{"current_steps": 277, "total_steps": 339, "loss": 0.0001, "lr": 8.02865739544767e-06, "epoch": 2.4500827357970216, "percentage": 81.71, "elapsed_time": "2:00:48", "remaining_time": "0:27:02", "throughput": 7892.41, "total_tokens": 57207824}
+{"current_steps": 278, "total_steps": 339, "loss": 0.0001, "lr": 7.778638817240042e-06, "epoch": 2.458907887479316, "percentage": 82.01, "elapsed_time": "2:01:16", "remaining_time": "0:26:36", "throughput": 7889.96, "total_tokens": 57415152}
+{"current_steps": 279, "total_steps": 339, "loss": 0.0001, "lr": 7.532246252988617e-06, "epoch": 2.467733039161611, "percentage": 82.3, "elapsed_time": "2:01:41", "remaining_time": "0:26:10", "throughput": 7892.18, "total_tokens": 57628096}
+{"current_steps": 280, "total_steps": 339, "loss": 0.0001, "lr": 7.289500863140414e-06, "epoch": 2.476558190843905, "percentage": 82.6, "elapsed_time": "2:02:05", "remaining_time": "0:25:43", "throughput": 7893.12, "total_tokens": 57824064}
+{"current_steps": 281, "total_steps": 339, "loss": 0.0002, "lr": 7.05042349491935e-06, "epoch": 2.4853833425261995, "percentage": 82.89, "elapsed_time": "2:02:35", "remaining_time": "0:25:18", "throughput": 7891.37, "total_tokens": 58042720}
+{"current_steps": 282, "total_steps": 339, "loss": 0.0001, "lr": 6.815034680535915e-06, "epoch": 2.494208494208494, "percentage": 83.19, "elapsed_time": "2:03:02", "remaining_time": "0:24:52", "throughput": 7891.23, "total_tokens": 58255408}
+{"current_steps": 283, "total_steps": 339, "loss": 0.0001, "lr": 6.5833546354237556e-06, "epoch": 2.5030336458907887, "percentage": 83.48, "elapsed_time": "2:03:29", "remaining_time": "0:24:26", "throughput": 7890.21, "total_tokens": 58464800}
+{"current_steps": 284, "total_steps": 339, "loss": 0.0001, "lr": 6.355403256503595e-06, "epoch": 2.5118587975730833, "percentage": 83.78, "elapsed_time": "2:03:56", "remaining_time": "0:24:00", "throughput": 7889.96, "total_tokens": 58672496}
+{"current_steps": 285, "total_steps": 339, "loss": 0.0002, "lr": 6.1312001204745115e-06, "epoch": 2.520683949255378, "percentage": 84.07, "elapsed_time": "2:04:25", "remaining_time": "0:23:34", "throughput": 7889.86, "total_tokens": 58898256}
+{"current_steps": 286, "total_steps": 339, "loss": 0.0001, "lr": 5.910764482132575e-06, "epoch": 2.5295091009376725, "percentage": 84.37, "elapsed_time": "2:04:51", "remaining_time": "0:23:08", "throughput": 7889.91, "total_tokens": 59107152}
+{"current_steps": 287, "total_steps": 339, "loss": 0.0002, "lr": 5.6941152727173265e-06, "epoch": 2.538334252619967, "percentage": 84.66, "elapsed_time": "2:05:15", "remaining_time": "0:22:41", "throughput": 7891.05, "total_tokens": 59307664}
+{"current_steps": 288, "total_steps": 339, "loss": 0.0003, "lr": 5.481271098285817e-06, "epoch": 2.5471594043022616, "percentage": 84.96, "elapsed_time": "2:05:42", "remaining_time": "0:22:15", "throughput": 7890.97, "total_tokens": 59514736}
+{"current_steps": 289, "total_steps": 339, "loss": 0.0001, "lr": 5.272250238114856e-06, "epoch": 2.5559845559845558, "percentage": 85.25, "elapsed_time": "2:06:06", "remaining_time": "0:21:49", "throughput": 7891.95, "total_tokens": 59712512}
+{"current_steps": 290, "total_steps": 339, "loss": 0.0001, "lr": 5.067070643131055e-06, "epoch": 2.564809707666851, "percentage": 85.55, "elapsed_time": "2:06:29", "remaining_time": "0:21:22", "throughput": 7893.32, "total_tokens": 59910000}
+{"current_steps": 291, "total_steps": 339, "loss": 0.0002, "lr": 4.865749934369223e-06, "epoch": 2.573634859349145, "percentage": 85.84, "elapsed_time": "2:06:55", "remaining_time": "0:20:56", "throughput": 7893.98, "total_tokens": 60116400}
+{"current_steps": 292, "total_steps": 339, "loss": 0.0002, "lr": 4.668305401459022e-06, "epoch": 2.5824600110314395, "percentage": 86.14, "elapsed_time": "2:07:22", "remaining_time": "0:20:30", "throughput": 7892.9, "total_tokens": 60320368}
+{"current_steps": 293, "total_steps": 339, "loss": 0.0001, "lr": 4.474754001140191e-06, "epoch": 2.591285162713734, "percentage": 86.43, "elapsed_time": "2:07:50", "remaining_time": "0:20:04", "throughput": 7892.36, "total_tokens": 60536528}
+{"current_steps": 294, "total_steps": 339, "loss": 0.001, "lr": 4.285112355806192e-06, "epoch": 2.6001103143960287, "percentage": 86.73, "elapsed_time": "2:08:17", "remaining_time": "0:19:38", "throughput": 7891.4, "total_tokens": 60743120}
+{"current_steps": 295, "total_steps": 339, "loss": 0.0001, "lr": 4.099396752076745e-06, "epoch": 2.6089354660783233, "percentage": 87.02, "elapsed_time": "2:08:42", "remaining_time": "0:19:11", "throughput": 7891.16, "total_tokens": 60942704}
+{"current_steps": 296, "total_steps": 339, "loss": 0.0001, "lr": 3.917623139399018e-06, "epoch": 2.617760617760618, "percentage": 87.32, "elapsed_time": "2:09:07", "remaining_time": "0:18:45", "throughput": 7892.02, "total_tokens": 61140128}
+{"current_steps": 297, "total_steps": 339, "loss": 0.0001, "lr": 3.7398071286779857e-06, "epoch": 2.6265857694429124, "percentage": 87.61, "elapsed_time": "2:09:30", "remaining_time": "0:18:18", "throughput": 7892.85, "total_tokens": 61334224}
+{"current_steps": 298, "total_steps": 339, "loss": 0.0001, "lr": 3.5659639909356723e-06, "epoch": 2.6354109211252066, "percentage": 87.91, "elapsed_time": "2:09:57", "remaining_time": "0:17:52", "throughput": 7892.72, "total_tokens": 61543280}
+{"current_steps": 299, "total_steps": 339, "loss": 0.0002, "lr": 3.3961086559996803e-06, "epoch": 2.6442360728075016, "percentage": 88.2, "elapsed_time": "2:10:23", "remaining_time": "0:17:26", "throughput": 7893.04, "total_tokens": 61750720}
+{"current_steps": 300, "total_steps": 339, "loss": 0.0, "lr": 3.230255711220992e-06, "epoch": 2.6530612244897958, "percentage": 88.5, "elapsed_time": "2:10:48", "remaining_time": "0:17:00", "throughput": 7892.55, "total_tokens": 61945952}
+{"current_steps": 301, "total_steps": 339, "loss": 0.0001, "lr": 3.0684194002212287e-06, "epoch": 2.6618863761720903, "percentage": 88.79, "elapsed_time": "2:11:14", "remaining_time": "0:16:34", "throughput": 7893.7, "total_tokens": 62155632}
+{"current_steps": 302, "total_steps": 339, "loss": 0.0001, "lr": 2.910613621669356e-06, "epoch": 2.670711527854385, "percentage": 89.09, "elapsed_time": "2:11:39", "remaining_time": "0:16:07", "throughput": 7893.69, "total_tokens": 62353216}
+{"current_steps": 303, "total_steps": 339, "loss": 0.0001, "lr": 2.7568519280880558e-06, "epoch": 2.6795366795366795, "percentage": 89.38, "elapsed_time": "2:12:03", "remaining_time": "0:15:41", "throughput": 7893.79, "total_tokens": 62544128}
+{"current_steps": 304, "total_steps": 339, "loss": 0.0004, "lr": 2.607147524689829e-06, "epoch": 2.688361831218974, "percentage": 89.68, "elapsed_time": "2:12:29", "remaining_time": "0:15:15", "throughput": 7893.85, "total_tokens": 62752688}
+{"current_steps": 305, "total_steps": 339, "loss": 0.0001, "lr": 2.4615132682429374e-06, "epoch": 2.6971869829012687, "percentage": 89.97, "elapsed_time": "2:12:58", "remaining_time": "0:14:49", "throughput": 7891.77, "total_tokens": 62963296}
+{"current_steps": 306, "total_steps": 339, "loss": 0.0002, "lr": 2.3199616659672354e-06, "epoch": 2.7060121345835633, "percentage": 90.27, "elapsed_time": "2:13:23", "remaining_time": "0:14:23", "throughput": 7892.22, "total_tokens": 63161904}
+{"current_steps": 307, "total_steps": 339, "loss": 0.0001, "lr": 2.182504874460006e-06, "epoch": 2.714837286265858, "percentage": 90.56, "elapsed_time": "2:13:49", "remaining_time": "0:13:56", "throughput": 7891.29, "total_tokens": 63365744}
+{"current_steps": 308, "total_steps": 339, "loss": 0.0001, "lr": 2.049154698651989e-06, "epoch": 2.7236624379481524, "percentage": 90.86, "elapsed_time": "2:14:14", "remaining_time": "0:13:30", "throughput": 7892.66, "total_tokens": 63571808}
+{"current_steps": 309, "total_steps": 339, "loss": 0.0001, "lr": 1.919922590793549e-06, "epoch": 2.7324875896304466, "percentage": 91.15, "elapsed_time": "2:14:39", "remaining_time": "0:13:04", "throughput": 7892.67, "total_tokens": 63768960}
+{"current_steps": 310, "total_steps": 339, "loss": 0.0001, "lr": 1.7948196494711188e-06, "epoch": 2.741312741312741, "percentage": 91.45, "elapsed_time": "2:15:05", "remaining_time": "0:12:38", "throughput": 7892.91, "total_tokens": 63979648}
+{"current_steps": 311, "total_steps": 339, "loss": 0.0001, "lr": 1.6738566186540627e-06, "epoch": 2.7501378929950357, "percentage": 91.74, "elapsed_time": "2:15:32", "remaining_time": "0:12:12", "throughput": 7892.87, "total_tokens": 64189712}
+{"current_steps": 312, "total_steps": 339, "loss": 0.0001, "lr": 1.5570438867719694e-06, "epoch": 2.7589630446773303, "percentage": 92.04, "elapsed_time": "2:15:59", "remaining_time": "0:11:46", "throughput": 7892.66, "total_tokens": 64400624}
+{"current_steps": 313, "total_steps": 339, "loss": 0.0001, "lr": 1.4443914858224938e-06, "epoch": 2.767788196359625, "percentage": 92.33, "elapsed_time": "2:16:26", "remaining_time": "0:11:20", "throughput": 7894.09, "total_tokens": 64626320}
+{"current_steps": 314, "total_steps": 339, "loss": 0.0001, "lr": 1.3359090905097848e-06, "epoch": 2.7766133480419195, "percentage": 92.63, "elapsed_time": "2:16:51", "remaining_time": "0:10:53", "throughput": 7894.56, "total_tokens": 64826480}
+{"current_steps": 315, "total_steps": 339, "loss": 0.0003, "lr": 1.2316060174136002e-06, "epoch": 2.785438499724214, "percentage": 92.92, "elapsed_time": "2:17:18", "remaining_time": "0:10:27", "throughput": 7893.92, "total_tokens": 65031984}
+{"current_steps": 316, "total_steps": 339, "loss": 0.0001, "lr": 1.1314912241892183e-06, "epoch": 2.7942636514065087, "percentage": 93.22, "elapsed_time": "2:17:43", "remaining_time": "0:10:01", "throughput": 7894.65, "total_tokens": 65239456}
+{"current_steps": 317, "total_steps": 339, "loss": 0.0002, "lr": 1.0355733087981378e-06, "epoch": 2.8030888030888033, "percentage": 93.51, "elapsed_time": "2:18:09", "remaining_time": "0:09:35", "throughput": 7893.46, "total_tokens": 65433888}
+{"current_steps": 318, "total_steps": 339, "loss": 0.0001, "lr": 9.43860508769645e-07, "epoch": 2.8119139547710974, "percentage": 93.81, "elapsed_time": "2:18:33", "remaining_time": "0:09:09", "throughput": 7894.99, "total_tokens": 65638288}
+{"current_steps": 319, "total_steps": 339, "loss": 0.0002, "lr": 8.563607004934193e-07, "epoch": 2.8207391064533924, "percentage": 94.1, "elapsed_time": "2:19:02", "remaining_time": "0:08:43", "throughput": 7894.0, "total_tokens": 65855952}
+{"current_steps": 320, "total_steps": 339, "loss": 0.0002, "lr": 7.730813985430407e-07, "epoch": 2.8295642581356866, "percentage": 94.4, "elapsed_time": "2:19:28", "remaining_time": "0:08:16", "throughput": 7895.06, "total_tokens": 66070192}
+{"current_steps": 321, "total_steps": 339, "loss": 0.0001, "lr": 6.940297550306896e-07, "epoch": 2.838389409817981, "percentage": 94.69, "elapsed_time": "2:19:56", "remaining_time": "0:07:50", "throughput": 7894.58, "total_tokens": 66283808}
+{"current_steps": 322, "total_steps": 339, "loss": 0.0002, "lr": 6.192125589928821e-07, "epoch": 2.8472145615002757, "percentage": 94.99, "elapsed_time": "2:20:25", "remaining_time": "0:07:24", "throughput": 7893.93, "total_tokens": 66507776}
+{"current_steps": 323, "total_steps": 339, "loss": 0.0002, "lr": 5.486362358074094e-07, "epoch": 2.8560397131825703, "percentage": 95.28, "elapsed_time": "2:20:50", "remaining_time": "0:06:58", "throughput": 7893.63, "total_tokens": 66708320}
+{"current_steps": 324, "total_steps": 339, "loss": 0.0001, "lr": 4.823068466415615e-07, "epoch": 2.864864864864865, "percentage": 95.58, "elapsed_time": "2:21:16", "remaining_time": "0:06:32", "throughput": 7893.28, "total_tokens": 66910032}
+{"current_steps": 325, "total_steps": 339, "loss": 0.0001, "lr": 4.202300879315446e-07, "epoch": 2.8736900165471595, "percentage": 95.87, "elapsed_time": "2:21:41", "remaining_time": "0:06:06", "throughput": 7894.28, "total_tokens": 67112784}
+{"current_steps": 326, "total_steps": 339, "loss": 0.0001, "lr": 3.624112908932942e-07, "epoch": 2.882515168229454, "percentage": 96.17, "elapsed_time": "2:22:05", "remaining_time": "0:05:39", "throughput": 7895.04, "total_tokens": 67306464}
+{"current_steps": 327, "total_steps": 339, "loss": 0.0001, "lr": 3.088554210646133e-07, "epoch": 2.8913403199117482, "percentage": 96.46, "elapsed_time": "2:22:29", "remaining_time": "0:05:13", "throughput": 7896.03, "total_tokens": 67504720}
+{"current_steps": 328, "total_steps": 339, "loss": 0.0001, "lr": 2.595670778787196e-07, "epoch": 2.9001654715940433, "percentage": 96.76, "elapsed_time": "2:22:52", "remaining_time": "0:04:47", "throughput": 7896.57, "total_tokens": 67694048}
+{"current_steps": 329, "total_steps": 339, "loss": 0.0002, "lr": 2.1455049426926666e-07, "epoch": 2.9089906232763374, "percentage": 97.05, "elapsed_time": "2:23:16", "remaining_time": "0:04:21", "throughput": 7897.7, "total_tokens": 67895008}
+{"current_steps": 330, "total_steps": 339, "loss": 0.0001, "lr": 1.7380953630678488e-07, "epoch": 2.917815774958632, "percentage": 97.35, "elapsed_time": "2:23:41", "remaining_time": "0:03:55", "throughput": 7898.05, "total_tokens": 68093168}
+{"current_steps": 331, "total_steps": 339, "loss": 0.0002, "lr": 1.373477028666803e-07, "epoch": 2.9266409266409266, "percentage": 97.64, "elapsed_time": "2:24:08", "remaining_time": "0:03:29", "throughput": 7898.34, "total_tokens": 68305568}
+{"current_steps": 332, "total_steps": 339, "loss": 0.0001, "lr": 1.0516812532873621e-07, "epoch": 2.935466078323221, "percentage": 97.94, "elapsed_time": "2:24:32", "remaining_time": "0:03:02", "throughput": 7898.84, "total_tokens": 68506384}
+{"current_steps": 333, "total_steps": 339, "loss": 0.0002, "lr": 7.727356730820035e-08, "epoch": 2.9442912300055157, "percentage": 98.23, "elapsed_time": "2:24:57", "remaining_time": "0:02:36", "throughput": 7900.67, "total_tokens": 68716160}
+{"current_steps": 334, "total_steps": 339, "loss": 0.0001, "lr": 5.3666424418413744e-08, "epoch": 2.9531163816878103, "percentage": 98.53, "elapsed_time": "2:25:22", "remaining_time": "0:02:10", "throughput": 7901.22, "total_tokens": 68918048}
+{"current_steps": 335, "total_steps": 339, "loss": 0.0001, "lr": 3.4348724065119685e-08, "epoch": 2.961941533370105, "percentage": 98.82, "elapsed_time": "2:25:50", "remaining_time": "0:01:44", "throughput": 7899.73, "total_tokens": 69129152}
+{"current_steps": 336, "total_steps": 339, "loss": 0.0003, "lr": 1.9322125272297488e-08, "epoch": 2.9707666850523995, "percentage": 99.12, "elapsed_time": "2:26:15", "remaining_time": "0:01:18", "throughput": 7900.03, "total_tokens": 69328576}
+{"current_steps": 337, "total_steps": 339, "loss": 0.0001, "lr": 8.587918539726402e-09, "epoch": 2.979591836734694, "percentage": 99.41, "elapsed_time": "2:26:39", "remaining_time": "0:00:52", "throughput": 7902.03, "total_tokens": 69537232}
+{"current_steps": 338, "total_steps": 339, "loss": 0.0001, "lr": 2.1470257321298813e-09, "epoch": 2.988416988416988, "percentage": 99.71, "elapsed_time": "2:27:06", "remaining_time": "0:00:26", "throughput": 7903.27, "total_tokens": 69761008}
+{"current_steps": 339, "total_steps": 339, "loss": 0.0001, "lr": 0.0, "epoch": 2.997242140099283, "percentage": 100.0, "elapsed_time": "2:27:31", "remaining_time": "0:00:00", "throughput": 7902.86, "total_tokens": 69953200}
+{"current_steps": 339, "total_steps": 339, "epoch": 2.997242140099283, "percentage": 100.0, "elapsed_time": "2:27:32", "remaining_time": "0:00:00", "throughput": 7902.18, "total_tokens": 69953200}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/trainer_state.json

@@ -0,0 +1,2755 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.997242140099283,
+  "eval_steps": 500,
+  "global_step": 339,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00882515168229454,
+      "grad_norm": 0.40674829483032227,
+      "learning_rate": 9.999785297426788e-05,
+      "loss": 0.2055,
+      "num_input_tokens_seen": 203120,
+      "step": 1
+    },
+    {
+      "epoch": 0.01765030336458908,
+      "grad_norm": 0.4242195785045624,
+      "learning_rate": 9.999141208146028e-05,
+      "loss": 0.1902,
+      "num_input_tokens_seen": 406048,
+      "step": 2
+    },
+    {
+      "epoch": 0.026475455046883617,
+      "grad_norm": 0.3813261389732361,
+      "learning_rate": 9.998067787472772e-05,
+      "loss": 0.1421,
+      "num_input_tokens_seen": 614736,
+      "step": 3
+    },
+    {
+      "epoch": 0.03530060672917816,
+      "grad_norm": 0.28003761172294617,
+      "learning_rate": 9.996565127593488e-05,
+      "loss": 0.1102,
+      "num_input_tokens_seen": 816416,
+      "step": 4
+    },
+    {
+      "epoch": 0.0441257584114727,
+      "grad_norm": 0.25300610065460205,
+      "learning_rate": 9.994633357558158e-05,
+      "loss": 0.0801,
+      "num_input_tokens_seen": 1024272,
+      "step": 5
+    },
+    {
+      "epoch": 0.052950910093767234,
+      "grad_norm": 0.2328871786594391,
+      "learning_rate": 9.99227264326918e-05,
+      "loss": 0.0574,
+      "num_input_tokens_seen": 1228192,
+      "step": 6
+    },
+    {
+      "epoch": 0.06177606177606178,
+      "grad_norm": 0.17362241446971893,
+      "learning_rate": 9.989483187467127e-05,
+      "loss": 0.0401,
+      "num_input_tokens_seen": 1434992,
+      "step": 7
+    },
+    {
+      "epoch": 0.07060121345835632,
+      "grad_norm": 0.09250874817371368,
+      "learning_rate": 9.986265229713331e-05,
+      "loss": 0.0295,
+      "num_input_tokens_seen": 1646560,
+      "step": 8
+    },
+    {
+      "epoch": 0.07942636514065085,
+      "grad_norm": 0.08936059474945068,
+      "learning_rate": 9.982619046369321e-05,
+      "loss": 0.0262,
+      "num_input_tokens_seen": 1838624,
+      "step": 9
+    },
+    {
+      "epoch": 0.0882515168229454,
+      "grad_norm": 0.08603595942258835,
+      "learning_rate": 9.978544950573074e-05,
+      "loss": 0.0263,
+      "num_input_tokens_seen": 2053488,
+      "step": 10
+    },
+    {
+      "epoch": 0.09707666850523994,
+      "grad_norm": 0.07848804444074631,
+      "learning_rate": 9.974043292212128e-05,
+      "loss": 0.022,
+      "num_input_tokens_seen": 2253680,
+      "step": 11
+    },
+    {
+      "epoch": 0.10590182018753447,
+      "grad_norm": 0.06246768683195114,
+      "learning_rate": 9.96911445789354e-05,
+      "loss": 0.0202,
+      "num_input_tokens_seen": 2442000,
+      "step": 12
+    },
+    {
+      "epoch": 0.11472697186982901,
+      "grad_norm": 0.048259809613227844,
+      "learning_rate": 9.963758870910671e-05,
+      "loss": 0.0202,
+      "num_input_tokens_seen": 2655920,
+      "step": 13
+    },
+    {
+      "epoch": 0.12355212355212356,
+      "grad_norm": 0.03917853534221649,
+      "learning_rate": 9.957976991206846e-05,
+      "loss": 0.0178,
+      "num_input_tokens_seen": 2874064,
+      "step": 14
+    },
+    {
+      "epoch": 0.13237727523441808,
+      "grad_norm": 0.040510393679142,
+      "learning_rate": 9.951769315335844e-05,
+      "loss": 0.0158,
+      "num_input_tokens_seen": 3071744,
+      "step": 15
+    },
+    {
+      "epoch": 0.14120242691671264,
+      "grad_norm": 0.035558607429265976,
+      "learning_rate": 9.945136376419259e-05,
+      "loss": 0.0159,
+      "num_input_tokens_seen": 3277904,
+      "step": 16
+    },
+    {
+      "epoch": 0.15002757859900717,
+      "grad_norm": 0.034995947033166885,
+      "learning_rate": 9.938078744100712e-05,
+      "loss": 0.0147,
+      "num_input_tokens_seen": 3493136,
+      "step": 17
+    },
+    {
+      "epoch": 0.1588527302813017,
+      "grad_norm": 0.03230876475572586,
+      "learning_rate": 9.930597024496931e-05,
+      "loss": 0.0138,
+      "num_input_tokens_seen": 3704288,
+      "step": 18
+    },
+    {
+      "epoch": 0.16767788196359626,
+      "grad_norm": 0.028281500563025475,
+      "learning_rate": 9.922691860145696e-05,
+      "loss": 0.0128,
+      "num_input_tokens_seen": 3904352,
+      "step": 19
+    },
+    {
+      "epoch": 0.1765030336458908,
+      "grad_norm": 0.026264235377311707,
+      "learning_rate": 9.914363929950659e-05,
+      "loss": 0.0124,
+      "num_input_tokens_seen": 4113888,
+      "step": 20
+    },
+    {
+      "epoch": 0.18532818532818532,
+      "grad_norm": 0.023232094943523407,
+      "learning_rate": 9.905613949123036e-05,
+      "loss": 0.0116,
+      "num_input_tokens_seen": 4323504,
+      "step": 21
+    },
+    {
+      "epoch": 0.19415333701047988,
+      "grad_norm": 0.02393435873091221,
+      "learning_rate": 9.896442669120187e-05,
+      "loss": 0.0109,
+      "num_input_tokens_seen": 4523008,
+      "step": 22
+    },
+    {
+      "epoch": 0.2029784886927744,
+      "grad_norm": 0.024421676993370056,
+      "learning_rate": 9.886850877581079e-05,
+      "loss": 0.0106,
+      "num_input_tokens_seen": 4732864,
+      "step": 23
+    },
+    {
+      "epoch": 0.21180364037506894,
+      "grad_norm": 0.022869078442454338,
+      "learning_rate": 9.876839398258641e-05,
+      "loss": 0.0099,
+      "num_input_tokens_seen": 4941936,
+      "step": 24
+    },
+    {
+      "epoch": 0.2206287920573635,
+      "grad_norm": 0.025933578610420227,
+      "learning_rate": 9.866409090949022e-05,
+      "loss": 0.0109,
+      "num_input_tokens_seen": 5143584,
+      "step": 25
+    },
+    {
+      "epoch": 0.22945394373965802,
+      "grad_norm": 0.02043001353740692,
+      "learning_rate": 9.855560851417752e-05,
+      "loss": 0.0084,
+      "num_input_tokens_seen": 5351024,
+      "step": 26
+    },
+    {
+      "epoch": 0.23827909542195255,
+      "grad_norm": 0.02140035293996334,
+      "learning_rate": 9.844295611322804e-05,
+      "loss": 0.0081,
+      "num_input_tokens_seen": 5563760,
+      "step": 27
+    },
+    {
+      "epoch": 0.2471042471042471,
+      "grad_norm": 0.019948888570070267,
+      "learning_rate": 9.832614338134595e-05,
+      "loss": 0.0078,
+      "num_input_tokens_seen": 5772416,
+      "step": 28
+    },
+    {
+      "epoch": 0.25592939878654164,
+      "grad_norm": 0.021153336390852928,
+      "learning_rate": 9.820518035052889e-05,
+      "loss": 0.0081,
+      "num_input_tokens_seen": 5974464,
+      "step": 29
+    },
+    {
+      "epoch": 0.26475455046883617,
+      "grad_norm": 0.02002059668302536,
+      "learning_rate": 9.808007740920646e-05,
+      "loss": 0.0087,
+      "num_input_tokens_seen": 6193520,
+      "step": 30
+    },
+    {
+      "epoch": 0.2735797021511307,
+      "grad_norm": 0.029256833717226982,
+      "learning_rate": 9.795084530134801e-05,
+      "loss": 0.0079,
+      "num_input_tokens_seen": 6399792,
+      "step": 31
+    },
+    {
+      "epoch": 0.2824048538334253,
+      "grad_norm": 0.02395695447921753,
+      "learning_rate": 9.781749512553999e-05,
+      "loss": 0.0086,
+      "num_input_tokens_seen": 6603584,
+      "step": 32
+    },
+    {
+      "epoch": 0.2912300055157198,
+      "grad_norm": 0.02185678854584694,
+      "learning_rate": 9.768003833403278e-05,
+      "loss": 0.0079,
+      "num_input_tokens_seen": 6810656,
+      "step": 33
+    },
+    {
+      "epoch": 0.30005515719801434,
+      "grad_norm": 0.02072463184595108,
+      "learning_rate": 9.753848673175707e-05,
+      "loss": 0.0069,
+      "num_input_tokens_seen": 7001792,
+      "step": 34
+    },
+    {
+      "epoch": 0.3088803088803089,
+      "grad_norm": 0.018024709075689316,
+      "learning_rate": 9.739285247531018e-05,
+      "loss": 0.0064,
+      "num_input_tokens_seen": 7205952,
+      "step": 35
+    },
+    {
+      "epoch": 0.3177054605626034,
+      "grad_norm": 0.019729286432266235,
+      "learning_rate": 9.724314807191195e-05,
+      "loss": 0.006,
+      "num_input_tokens_seen": 7406304,
+      "step": 36
+    },
+    {
+      "epoch": 0.32653061224489793,
+      "grad_norm": 0.01830880530178547,
+      "learning_rate": 9.708938637833065e-05,
+      "loss": 0.0067,
+      "num_input_tokens_seen": 7629568,
+      "step": 37
+    },
+    {
+      "epoch": 0.3353557639271925,
+      "grad_norm": 0.021113887429237366,
+      "learning_rate": 9.693158059977878e-05,
+      "loss": 0.0063,
+      "num_input_tokens_seen": 7845200,
+      "step": 38
+    },
+    {
+      "epoch": 0.34418091560948705,
+      "grad_norm": 0.015138108283281326,
+      "learning_rate": 9.676974428877901e-05,
+      "loss": 0.0058,
+      "num_input_tokens_seen": 8061840,
+      "step": 39
+    },
+    {
+      "epoch": 0.3530060672917816,
+      "grad_norm": 0.017043087631464005,
+      "learning_rate": 9.660389134400033e-05,
+      "loss": 0.0061,
+      "num_input_tokens_seen": 8279664,
+      "step": 40
+    },
+    {
+      "epoch": 0.3618312189740761,
+      "grad_norm": 0.01955767348408699,
+      "learning_rate": 9.643403600906433e-05,
+      "loss": 0.0055,
+      "num_input_tokens_seen": 8475376,
+      "step": 41
+    },
+    {
+      "epoch": 0.37065637065637064,
+      "grad_norm": 0.014688636176288128,
+      "learning_rate": 9.626019287132203e-05,
+      "loss": 0.005,
+      "num_input_tokens_seen": 8691760,
+      "step": 42
+    },
+    {
+      "epoch": 0.3794815223386652,
+      "grad_norm": 0.01973150670528412,
+      "learning_rate": 9.608237686060099e-05,
+      "loss": 0.006,
+      "num_input_tokens_seen": 8884736,
+      "step": 43
+    },
+    {
+      "epoch": 0.38830667402095975,
+      "grad_norm": 0.01489401888102293,
+      "learning_rate": 9.590060324792327e-05,
+      "loss": 0.0048,
+      "num_input_tokens_seen": 9084064,
+      "step": 44
+    },
+    {
+      "epoch": 0.3971318257032543,
+      "grad_norm": 0.015995647758245468,
+      "learning_rate": 9.571488764419381e-05,
+      "loss": 0.0047,
+      "num_input_tokens_seen": 9302144,
+      "step": 45
+    },
+    {
+      "epoch": 0.4059569773855488,
+      "grad_norm": 0.01859475113451481,
+      "learning_rate": 9.552524599885981e-05,
+      "loss": 0.0053,
+      "num_input_tokens_seen": 9517456,
+      "step": 46
+    },
+    {
+      "epoch": 0.41478212906784334,
+      "grad_norm": 0.018746482208371162,
+      "learning_rate": 9.533169459854098e-05,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 9710768,
+      "step": 47
+    },
+    {
+      "epoch": 0.42360728075013787,
+      "grad_norm": 0.017155013978481293,
+      "learning_rate": 9.513425006563079e-05,
+      "loss": 0.0043,
+      "num_input_tokens_seen": 9914064,
+      "step": 48
+    },
+    {
+      "epoch": 0.43243243243243246,
+      "grad_norm": 0.015938682481646538,
+      "learning_rate": 9.493292935686895e-05,
+      "loss": 0.0041,
+      "num_input_tokens_seen": 10120208,
+      "step": 49
+    },
+    {
+      "epoch": 0.441257584114727,
+      "grad_norm": 0.017114240676164627,
+      "learning_rate": 9.472774976188515e-05,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 10346304,
+      "step": 50
+    },
+    {
+      "epoch": 0.4500827357970215,
+      "grad_norm": 0.014332287944853306,
+      "learning_rate": 9.451872890171419e-05,
+      "loss": 0.004,
+      "num_input_tokens_seen": 10547984,
+      "step": 51
+    },
+    {
+      "epoch": 0.45890788747931605,
+      "grad_norm": 0.017018554732203484,
+      "learning_rate": 9.43058847272827e-05,
+      "loss": 0.0045,
+      "num_input_tokens_seen": 10754288,
+      "step": 52
+    },
+    {
+      "epoch": 0.4677330391616106,
+      "grad_norm": 0.013670100830495358,
+      "learning_rate": 9.408923551786743e-05,
+      "loss": 0.0028,
+      "num_input_tokens_seen": 10942704,
+      "step": 53
+    },
+    {
+      "epoch": 0.4765581908439051,
+      "grad_norm": 0.016749229282140732,
+      "learning_rate": 9.386879987952549e-05,
+      "loss": 0.0034,
+      "num_input_tokens_seen": 11150864,
+      "step": 54
+    },
+    {
+      "epoch": 0.4853833425261997,
+      "grad_norm": 0.01554529182612896,
+      "learning_rate": 9.364459674349641e-05,
+      "loss": 0.0042,
+      "num_input_tokens_seen": 11367728,
+      "step": 55
+    },
+    {
+      "epoch": 0.4942084942084942,
+      "grad_norm": 0.015070905908942223,
+      "learning_rate": 9.341664536457626e-05,
+      "loss": 0.0028,
+      "num_input_tokens_seen": 11575536,
+      "step": 56
+    },
+    {
+      "epoch": 0.5030336458907887,
+      "grad_norm": 0.016440849751234055,
+      "learning_rate": 9.31849653194641e-05,
+      "loss": 0.0035,
+      "num_input_tokens_seen": 11781328,
+      "step": 57
+    },
+    {
+      "epoch": 0.5118587975730833,
+      "grad_norm": 0.014468475244939327,
+      "learning_rate": 9.294957650508065e-05,
+      "loss": 0.0029,
+      "num_input_tokens_seen": 11981232,
+      "step": 58
+    },
+    {
+      "epoch": 0.5206839492553779,
+      "grad_norm": 0.014588565565645695,
+      "learning_rate": 9.27104991368596e-05,
+      "loss": 0.0028,
+      "num_input_tokens_seen": 12187296,
+      "step": 59
+    },
+    {
+      "epoch": 0.5295091009376723,
+      "grad_norm": 0.0141281234100461,
+      "learning_rate": 9.246775374701139e-05,
+      "loss": 0.0027,
+      "num_input_tokens_seen": 12385632,
+      "step": 60
+    },
+    {
+      "epoch": 0.5383342526199669,
+      "grad_norm": 0.013463583774864674,
+      "learning_rate": 9.222136118275995e-05,
+      "loss": 0.0022,
+      "num_input_tokens_seen": 12588928,
+      "step": 61
+    },
+    {
+      "epoch": 0.5471594043022614,
+      "grad_norm": 0.014033553190529346,
+      "learning_rate": 9.197134260455233e-05,
+      "loss": 0.0027,
+      "num_input_tokens_seen": 12825616,
+      "step": 62
+    },
+    {
+      "epoch": 0.555984555984556,
+      "grad_norm": 0.013906535692512989,
+      "learning_rate": 9.171771948424137e-05,
+      "loss": 0.0025,
+      "num_input_tokens_seen": 13044976,
+      "step": 63
+    },
+    {
+      "epoch": 0.5648097076668506,
+      "grad_norm": 0.012418747879564762,
+      "learning_rate": 9.146051360324166e-05,
+      "loss": 0.0025,
+      "num_input_tokens_seen": 13255280,
+      "step": 64
+    },
+    {
+      "epoch": 0.573634859349145,
+      "grad_norm": 0.015126565471291542,
+      "learning_rate": 9.119974705065901e-05,
+      "loss": 0.0022,
+      "num_input_tokens_seen": 13463456,
+      "step": 65
+    },
+    {
+      "epoch": 0.5824600110314396,
+      "grad_norm": 0.013123284094035625,
+      "learning_rate": 9.093544222139337e-05,
+      "loss": 0.0023,
+      "num_input_tokens_seen": 13667744,
+      "step": 66
+    },
+    {
+      "epoch": 0.5912851627137341,
+      "grad_norm": 0.014246366918087006,
+      "learning_rate": 9.066762181421552e-05,
+      "loss": 0.0024,
+      "num_input_tokens_seen": 13874240,
+      "step": 67
+    },
+    {
+      "epoch": 0.6001103143960287,
+      "grad_norm": 0.011402356438338757,
+      "learning_rate": 9.039630882981768e-05,
+      "loss": 0.0015,
+      "num_input_tokens_seen": 14081392,
+      "step": 68
+    },
+    {
+      "epoch": 0.6089354660783233,
+      "grad_norm": 0.014725148677825928,
+      "learning_rate": 9.012152656883823e-05,
+      "loss": 0.0033,
+      "num_input_tokens_seen": 14300896,
+      "step": 69
+    },
+    {
+      "epoch": 0.6177606177606177,
+      "grad_norm": 0.014837515540421009,
+      "learning_rate": 8.984329862986056e-05,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 14523968,
+      "step": 70
+    },
+    {
+      "epoch": 0.6265857694429123,
+      "grad_norm": 0.014493652619421482,
+      "learning_rate": 8.956164890738643e-05,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 14728960,
+      "step": 71
+    },
+    {
+      "epoch": 0.6354109211252068,
+      "grad_norm": 0.011806878261268139,
+      "learning_rate": 8.927660158978392e-05,
+      "loss": 0.0016,
+      "num_input_tokens_seen": 14912480,
+      "step": 72
+    },
+    {
+      "epoch": 0.6442360728075014,
+      "grad_norm": 0.01818985864520073,
+      "learning_rate": 8.898818115721008e-05,
+      "loss": 0.0019,
+      "num_input_tokens_seen": 15114608,
+      "step": 73
+    },
+    {
+      "epoch": 0.6530612244897959,
+      "grad_norm": 0.015412255190312862,
+      "learning_rate": 8.86964123795085e-05,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 15326112,
+      "step": 74
+    },
+    {
+      "epoch": 0.6618863761720905,
+      "grad_norm": 0.013063928112387657,
+      "learning_rate": 8.84013203140821e-05,
+      "loss": 0.0015,
+      "num_input_tokens_seen": 15545248,
+      "step": 75
+    },
+    {
+      "epoch": 0.670711527854385,
+      "grad_norm": 0.016336796805262566,
+      "learning_rate": 8.810293030374126e-05,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 15751872,
+      "step": 76
+    },
+    {
+      "epoch": 0.6795366795366795,
+      "grad_norm": 0.010313590988516808,
+      "learning_rate": 8.780126797452713e-05,
+      "loss": 0.001,
+      "num_input_tokens_seen": 15957872,
+      "step": 77
+    },
+    {
+      "epoch": 0.6883618312189741,
+      "grad_norm": 0.015468253754079342,
+      "learning_rate": 8.749635923351107e-05,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 16162640,
+      "step": 78
+    },
+    {
+      "epoch": 0.6971869829012686,
+      "grad_norm": 0.01543041318655014,
+      "learning_rate": 8.71882302665696e-05,
+      "loss": 0.001,
+      "num_input_tokens_seen": 16352368,
+      "step": 79
+    },
+    {
+      "epoch": 0.7060121345835632,
+      "grad_norm": 0.01957864873111248,
+      "learning_rate": 8.687690753613554e-05,
+      "loss": 0.0014,
+      "num_input_tokens_seen": 16563920,
+      "step": 80
+    },
+    {
+      "epoch": 0.7148372862658577,
+      "grad_norm": 0.012508533895015717,
+      "learning_rate": 8.656241777892543e-05,
+      "loss": 0.001,
+      "num_input_tokens_seen": 16759024,
+      "step": 81
+    },
+    {
+      "epoch": 0.7236624379481522,
+      "grad_norm": 0.012273616157472134,
+      "learning_rate": 8.624478800364332e-05,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 16973728,
+      "step": 82
+    },
+    {
+      "epoch": 0.7324875896304468,
+      "grad_norm": 0.01503776852041483,
+      "learning_rate": 8.592404548866123e-05,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 17162752,
+      "step": 83
+    },
+    {
+      "epoch": 0.7413127413127413,
+      "grad_norm": 0.014227951876819134,
+      "learning_rate": 8.560021777967649e-05,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 17364064,
+      "step": 84
+    },
+    {
+      "epoch": 0.7501378929950359,
+      "grad_norm": 0.01252016518265009,
+      "learning_rate": 8.527333268734606e-05,
+      "loss": 0.0011,
+      "num_input_tokens_seen": 17564576,
+      "step": 85
+    },
+    {
+      "epoch": 0.7589630446773304,
+      "grad_norm": 0.011520475149154663,
+      "learning_rate": 8.494341828489812e-05,
+      "loss": 0.0037,
+      "num_input_tokens_seen": 17778752,
+      "step": 86
+    },
+    {
+      "epoch": 0.7677881963596249,
+      "grad_norm": 0.010531144216656685,
+      "learning_rate": 8.461050290572114e-05,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 17982448,
+      "step": 87
+    },
+    {
+      "epoch": 0.7766133480419195,
+      "grad_norm": 0.010875461623072624,
+      "learning_rate": 8.427461514093056e-05,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 18180608,
+      "step": 88
+    },
+    {
+      "epoch": 0.785438499724214,
+      "grad_norm": 0.007611530367285013,
+      "learning_rate": 8.393578383691329e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 18384496,
+      "step": 89
+    },
+    {
+      "epoch": 0.7942636514065086,
+      "grad_norm": 0.010159923695027828,
+      "learning_rate": 8.359403809285053e-05,
+      "loss": 0.001,
+      "num_input_tokens_seen": 18587744,
+      "step": 90
+    },
+    {
+      "epoch": 0.803088803088803,
+      "grad_norm": 0.011715343222022057,
+      "learning_rate": 8.324940725821852e-05,
+      "loss": 0.001,
+      "num_input_tokens_seen": 18791056,
+      "step": 91
+    },
+    {
+      "epoch": 0.8119139547710976,
+      "grad_norm": 0.012972251512110233,
+      "learning_rate": 8.290192093026805e-05,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 18985008,
+      "step": 92
+    },
+    {
+      "epoch": 0.8207391064533922,
+      "grad_norm": 0.0135871022939682,
+      "learning_rate": 8.255160895148263e-05,
+      "loss": 0.0014,
+      "num_input_tokens_seen": 19193888,
+      "step": 93
+    },
+    {
+      "epoch": 0.8295642581356867,
+      "grad_norm": 0.011914449743926525,
+      "learning_rate": 8.219850140701557e-05,
+      "loss": 0.001,
+      "num_input_tokens_seen": 19399552,
+      "step": 94
+    },
+    {
+      "epoch": 0.8383894098179813,
+      "grad_norm": 0.009591113775968552,
+      "learning_rate": 8.184262862210624e-05,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 19605120,
+      "step": 95
+    },
+    {
+      "epoch": 0.8472145615002757,
+      "grad_norm": 0.009942690841853619,
+      "learning_rate": 8.148402115947571e-05,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 19802480,
+      "step": 96
+    },
+    {
+      "epoch": 0.8560397131825703,
+      "grad_norm": 0.012667879462242126,
+      "learning_rate": 8.112270981670196e-05,
+      "loss": 0.0011,
+      "num_input_tokens_seen": 20009520,
+      "step": 97
+    },
+    {
+      "epoch": 0.8648648648648649,
+      "grad_norm": 0.010983509942889214,
+      "learning_rate": 8.075872562357501e-05,
+      "loss": 0.0009,
+      "num_input_tokens_seen": 20235888,
+      "step": 98
+    },
+    {
+      "epoch": 0.8736900165471594,
+      "grad_norm": 0.011479397304356098,
+      "learning_rate": 8.039209983943201e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 20433600,
+      "step": 99
+    },
+    {
+      "epoch": 0.882515168229454,
+      "grad_norm": 0.012184002436697483,
+      "learning_rate": 8.002286395047267e-05,
+      "loss": 0.0009,
+      "num_input_tokens_seen": 20631664,
+      "step": 100
+    },
+    {
+      "epoch": 0.8913403199117484,
+      "grad_norm": 0.009395604953169823,
+      "learning_rate": 7.965104966705518e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 20833056,
+      "step": 101
+    },
+    {
+      "epoch": 0.900165471594043,
+      "grad_norm": 0.013585143722593784,
+      "learning_rate": 7.927668892097289e-05,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 21051104,
+      "step": 102
+    },
+    {
+      "epoch": 0.9089906232763376,
+      "grad_norm": 0.008882119320333004,
+      "learning_rate": 7.889981386271201e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 21246080,
+      "step": 103
+    },
+    {
+      "epoch": 0.9178157749586321,
+      "grad_norm": 0.010433576069772243,
+      "learning_rate": 7.852045685869045e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 21439696,
+      "step": 104
+    },
+    {
+      "epoch": 0.9266409266409267,
+      "grad_norm": 0.01474383007735014,
+      "learning_rate": 7.813865048847819e-05,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 21648432,
+      "step": 105
+    },
+    {
+      "epoch": 0.9354660783232212,
+      "grad_norm": 0.011113091371953487,
+      "learning_rate": 7.775442754199928e-05,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 21864368,
+      "step": 106
+    },
+    {
+      "epoch": 0.9442912300055157,
+      "grad_norm": 0.009181715548038483,
+      "learning_rate": 7.736782101671587e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 22061968,
+      "step": 107
+    },
+    {
+      "epoch": 0.9531163816878102,
+      "grad_norm": 0.0140100521966815,
+      "learning_rate": 7.697886411479423e-05,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 22278128,
+      "step": 108
+    },
+    {
+      "epoch": 0.9619415333701048,
+      "grad_norm": 0.007349591236561537,
+      "learning_rate": 7.658759024025349e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 22469056,
+      "step": 109
+    },
+    {
+      "epoch": 0.9707666850523994,
+      "grad_norm": 0.01252900529652834,
+      "learning_rate": 7.619403299609668e-05,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 22662128,
+      "step": 110
+    },
+    {
+      "epoch": 0.9795918367346939,
+      "grad_norm": 0.012083148583769798,
+      "learning_rate": 7.579822618142505e-05,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 22883216,
+      "step": 111
+    },
+    {
+      "epoch": 0.9884169884169884,
+      "grad_norm": 0.010517132468521595,
+      "learning_rate": 7.540020378853523e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 23085888,
+      "step": 112
+    },
+    {
+      "epoch": 0.9972421400992829,
+      "grad_norm": 0.01143716461956501,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 23307520,
+      "step": 113
+    },
+    {
+      "epoch": 1.0088251516822946,
+      "grad_norm": 0.0287212785333395,
+      "learning_rate": 7.459764918573264e-05,
+      "loss": 0.0014,
+      "num_input_tokens_seen": 23564192,
+      "step": 114
+    },
+    {
+      "epoch": 1.0176503033645892,
+      "grad_norm": 0.010353313758969307,
+      "learning_rate": 7.419318590003523e-05,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 23768816,
+      "step": 115
+    },
+    {
+      "epoch": 1.0264754550468835,
+      "grad_norm": 0.013796573504805565,
+      "learning_rate": 7.378664487863103e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 23974096,
+      "step": 116
+    },
+    {
+      "epoch": 1.0353006067291781,
+      "grad_norm": 0.006352484691888094,
+      "learning_rate": 7.33780610356814e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 24172256,
+      "step": 117
+    },
+    {
+      "epoch": 1.0441257584114727,
+      "grad_norm": 0.007957457564771175,
+      "learning_rate": 7.296746946078736e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 24362208,
+      "step": 118
+    },
+    {
+      "epoch": 1.0529509100937673,
+      "grad_norm": 0.0068214968778193,
+      "learning_rate": 7.255490541597594e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 24562224,
+      "step": 119
+    },
+    {
+      "epoch": 1.0617760617760619,
+      "grad_norm": 0.00877879373729229,
+      "learning_rate": 7.214040433267198e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 24776528,
+      "step": 120
+    },
+    {
+      "epoch": 1.0706012134583562,
+      "grad_norm": 0.007200079504400492,
+      "learning_rate": 7.172400180865513e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 24985008,
+      "step": 121
+    },
+    {
+      "epoch": 1.0794263651406508,
+      "grad_norm": 0.010829208418726921,
+      "learning_rate": 7.130573360500276e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 25200720,
+      "step": 122
+    },
+    {
+      "epoch": 1.0882515168229454,
+      "grad_norm": 0.010170291177928448,
+      "learning_rate": 7.088563564301873e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 25413568,
+      "step": 123
+    },
+    {
+      "epoch": 1.09707666850524,
+      "grad_norm": 0.007032219786196947,
+      "learning_rate": 7.046374400114842e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 25608576,
+      "step": 124
+    },
+    {
+      "epoch": 1.1059018201875346,
+      "grad_norm": 0.00843306165188551,
+      "learning_rate": 7.004009491188022e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 25818400,
+      "step": 125
+    },
+    {
+      "epoch": 1.114726971869829,
+      "grad_norm": 0.00947788916528225,
+      "learning_rate": 6.961472475863405e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 26037424,
+      "step": 126
+    },
+    {
+      "epoch": 1.1235521235521235,
+      "grad_norm": 0.009593469090759754,
+      "learning_rate": 6.918767007263646e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 26250480,
+      "step": 127
+    },
+    {
+      "epoch": 1.1323772752344181,
+      "grad_norm": 0.012611499056220055,
+      "learning_rate": 6.875896752978344e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 26458592,
+      "step": 128
+    },
+    {
+      "epoch": 1.1412024269167127,
+      "grad_norm": 0.005860932637006044,
+      "learning_rate": 6.832865394749065e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 26680256,
+      "step": 129
+    },
+    {
+      "epoch": 1.150027578599007,
+      "grad_norm": 0.008905632421374321,
+      "learning_rate": 6.789676628153143e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 26887424,
+      "step": 130
+    },
+    {
+      "epoch": 1.1588527302813016,
+      "grad_norm": 0.00839240662753582,
+      "learning_rate": 6.746334162286307e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 27112736,
+      "step": 131
+    },
+    {
+      "epoch": 1.1676778819635962,
+      "grad_norm": 0.010829194448888302,
+      "learning_rate": 6.702841719444141e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 27320064,
+      "step": 132
+    },
+    {
+      "epoch": 1.1765030336458908,
+      "grad_norm": 0.005576102528721094,
+      "learning_rate": 6.659203034802397e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 27520544,
+      "step": 133
+    },
+    {
+      "epoch": 1.1853281853281854,
+      "grad_norm": 0.008609413169324398,
+      "learning_rate": 6.615421856096231e-05,
+      "loss": 0.0009,
+      "num_input_tokens_seen": 27737920,
+      "step": 134
+    },
+    {
+      "epoch": 1.19415333701048,
+      "grad_norm": 0.013195198960602283,
+      "learning_rate": 6.571501943298334e-05,
+      "loss": 0.0014,
+      "num_input_tokens_seen": 27947552,
+      "step": 135
+    },
+    {
+      "epoch": 1.2029784886927744,
+      "grad_norm": 0.008647961542010307,
+      "learning_rate": 6.527447068296026e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 28143808,
+      "step": 136
+    },
+    {
+      "epoch": 1.211803640375069,
+      "grad_norm": 0.006975845899432898,
+      "learning_rate": 6.483261014567311e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 28349312,
+      "step": 137
+    },
+    {
+      "epoch": 1.2206287920573635,
+      "grad_norm": 0.013750969432294369,
+      "learning_rate": 6.438947576855968e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 28560096,
+      "step": 138
+    },
+    {
+      "epoch": 1.229453943739658,
+      "grad_norm": 0.009799162857234478,
+      "learning_rate": 6.394510560845637e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 28764544,
+      "step": 139
+    },
+    {
+      "epoch": 1.2382790954219525,
+      "grad_norm": 0.00819414108991623,
+      "learning_rate": 6.349953782832991e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 28949360,
+      "step": 140
+    },
+    {
+      "epoch": 1.247104247104247,
+      "grad_norm": 0.008884673938155174,
+      "learning_rate": 6.305281069399989e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 29148112,
+      "step": 141
+    },
+    {
+      "epoch": 1.2559293987865416,
+      "grad_norm": 0.009248818270862103,
+      "learning_rate": 6.26049625708524e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 29370624,
+      "step": 142
+    },
+    {
+      "epoch": 1.2647545504688362,
+      "grad_norm": 0.008902438916265965,
+      "learning_rate": 6.215603192054522e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 29572464,
+      "step": 143
+    },
+    {
+      "epoch": 1.2735797021511308,
+      "grad_norm": 0.012439709156751633,
+      "learning_rate": 6.17060572977047e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 29771152,
+      "step": 144
+    },
+    {
+      "epoch": 1.2824048538334254,
+      "grad_norm": 0.013059360906481743,
+      "learning_rate": 6.125507734661458e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 29954960,
+      "step": 145
+    },
+    {
+      "epoch": 1.2912300055157198,
+      "grad_norm": 0.011295526288449764,
+      "learning_rate": 6.080313079789723e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 30165568,
+      "step": 146
+    },
+    {
+      "epoch": 1.3000551571980143,
+      "grad_norm": 0.01000818982720375,
+      "learning_rate": 6.035025646518746e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 30372160,
+      "step": 147
+    },
+    {
+      "epoch": 1.308880308880309,
+      "grad_norm": 0.010914387181401253,
+      "learning_rate": 5.989649324179911e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 30572752,
+      "step": 148
+    },
+    {
+      "epoch": 1.3177054605626033,
+      "grad_norm": 0.009289560839533806,
+      "learning_rate": 5.944188009738483e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 30780496,
+      "step": 149
+    },
+    {
+      "epoch": 1.3265306122448979,
+      "grad_norm": 0.015559184364974499,
+      "learning_rate": 5.8986456074589404e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 30975120,
+      "step": 150
+    },
+    {
+      "epoch": 1.3353557639271925,
+      "grad_norm": 0.00643413420766592,
+      "learning_rate": 5.853026028569667e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 31174000,
+      "step": 151
+    },
+    {
+      "epoch": 1.344180915609487,
+      "grad_norm": 0.0077626509591937065,
+      "learning_rate": 5.807333190927053e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 31387088,
+      "step": 152
+    },
+    {
+      "epoch": 1.3530060672917816,
+      "grad_norm": 0.0083751380443573,
+      "learning_rate": 5.761571018679025e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 31576400,
+      "step": 153
+    },
+    {
+      "epoch": 1.3618312189740762,
+      "grad_norm": 0.007961435243487358,
+      "learning_rate": 5.715743441928041e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 31784320,
+      "step": 154
+    },
+    {
+      "epoch": 1.3706563706563706,
+      "grad_norm": 0.006737589370459318,
+      "learning_rate": 5.669854396393559e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 31987520,
+      "step": 155
+    },
+    {
+      "epoch": 1.3794815223386652,
+      "grad_norm": 0.014642222784459591,
+      "learning_rate": 5.6239078230740436e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 32187456,
+      "step": 156
+    },
+    {
+      "epoch": 1.3883066740209598,
+      "grad_norm": 0.006064648274332285,
+      "learning_rate": 5.5779076679085054e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 32384528,
+      "step": 157
+    },
+    {
+      "epoch": 1.3971318257032543,
+      "grad_norm": 0.009461612440645695,
+      "learning_rate": 5.531857881437612e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 32593040,
+      "step": 158
+    },
+    {
+      "epoch": 1.4059569773855487,
+      "grad_norm": 0.007511747535318136,
+      "learning_rate": 5.48576241846443e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 32797952,
+      "step": 159
+    },
+    {
+      "epoch": 1.4147821290678433,
+      "grad_norm": 0.02702983096241951,
+      "learning_rate": 5.4396252377147615e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 33008800,
+      "step": 160
+    },
+    {
+      "epoch": 1.4236072807501379,
+      "grad_norm": 0.008439299650490284,
+      "learning_rate": 5.3934503014971793e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 33208352,
+      "step": 161
+    },
+    {
+      "epoch": 1.4324324324324325,
+      "grad_norm": 0.0037907836958765984,
+      "learning_rate": 5.347241575362729e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 33410208,
+      "step": 162
+    },
+    {
+      "epoch": 1.441257584114727,
+      "grad_norm": 0.008237862028181553,
+      "learning_rate": 5.30100302776438e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 33631888,
+      "step": 163
+    },
+    {
+      "epoch": 1.4500827357970216,
+      "grad_norm": 0.009860441088676453,
+      "learning_rate": 5.254738629716186e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 33825152,
+      "step": 164
+    },
+    {
+      "epoch": 1.458907887479316,
+      "grad_norm": 0.007564296945929527,
+      "learning_rate": 5.208452354452274e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 34020352,
+      "step": 165
+    },
+    {
+      "epoch": 1.4677330391616106,
+      "grad_norm": 0.019607344642281532,
+      "learning_rate": 5.162148177085604e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 34226288,
+      "step": 166
+    },
+    {
+      "epoch": 1.4765581908439052,
+      "grad_norm": 0.007924061268568039,
+      "learning_rate": 5.115830074266591e-05,
+      "loss": 0.0016,
+      "num_input_tokens_seen": 34426672,
+      "step": 167
+    },
+    {
+      "epoch": 1.4853833425261997,
+      "grad_norm": 0.006358864717185497,
+      "learning_rate": 5.0695020238415756e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 34636944,
+      "step": 168
+    },
+    {
+      "epoch": 1.494208494208494,
+      "grad_norm": 0.010681587271392345,
+      "learning_rate": 5.0231680045112176e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 34839456,
+      "step": 169
+    },
+    {
+      "epoch": 1.5030336458907887,
+      "grad_norm": 0.01033815648406744,
+      "learning_rate": 4.976831995488784e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 35031600,
+      "step": 170
+    },
+    {
+      "epoch": 1.5118587975730833,
+      "grad_norm": 0.016812577843666077,
+      "learning_rate": 4.9304979761584256e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 35227728,
+      "step": 171
+    },
+    {
+      "epoch": 1.5206839492553779,
+      "grad_norm": 0.008957776241004467,
+      "learning_rate": 4.884169925733409e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 35436528,
+      "step": 172
+    },
+    {
+      "epoch": 1.5295091009376725,
+      "grad_norm": 0.006675931625068188,
+      "learning_rate": 4.837851822914397e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 35628624,
+      "step": 173
+    },
+    {
+      "epoch": 1.538334252619967,
+      "grad_norm": 0.006146900821477175,
+      "learning_rate": 4.791547645547726e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 35827376,
+      "step": 174
+    },
+    {
+      "epoch": 1.5471594043022614,
+      "grad_norm": 0.012180755846202374,
+      "learning_rate": 4.745261370283817e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 36056560,
+      "step": 175
+    },
+    {
+      "epoch": 1.555984555984556,
+      "grad_norm": 0.00920344889163971,
+      "learning_rate": 4.698996972235622e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 36267568,
+      "step": 176
+    },
+    {
+      "epoch": 1.5648097076668506,
+      "grad_norm": 0.010103096254169941,
+      "learning_rate": 4.652758424637271e-05,
+      "loss": 0.0027,
+      "num_input_tokens_seen": 36473008,
+      "step": 177
+    },
+    {
+      "epoch": 1.573634859349145,
+      "grad_norm": 0.012086655013263226,
+      "learning_rate": 4.606549698502823e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 36670944,
+      "step": 178
+    },
+    {
+      "epoch": 1.5824600110314395,
+      "grad_norm": 0.0054108137264847755,
+      "learning_rate": 4.56037476228524e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 36882256,
+      "step": 179
+    },
+    {
+      "epoch": 1.591285162713734,
+      "grad_norm": 0.014871139079332352,
+      "learning_rate": 4.5142375815355706e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 37091392,
+      "step": 180
+    },
+    {
+      "epoch": 1.6001103143960287,
+      "grad_norm": 0.005915229208767414,
+      "learning_rate": 4.468142118562389e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 37309680,
+      "step": 181
+    },
+    {
+      "epoch": 1.6089354660783233,
+      "grad_norm": 0.006937643978744745,
+      "learning_rate": 4.4220923320914964e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 37517952,
+      "step": 182
+    },
+    {
+      "epoch": 1.6177606177606179,
+      "grad_norm": 0.00866376981139183,
+      "learning_rate": 4.376092176925958e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 37732160,
+      "step": 183
+    },
+    {
+      "epoch": 1.6265857694429124,
+      "grad_norm": 0.007841500453650951,
+      "learning_rate": 4.330145603606441e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 37940368,
+      "step": 184
+    },
+    {
+      "epoch": 1.6354109211252068,
+      "grad_norm": 0.008568421937525272,
+      "learning_rate": 4.2842565580719595e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 38135024,
+      "step": 185
+    },
+    {
+      "epoch": 1.6442360728075014,
+      "grad_norm": 0.011796732433140278,
+      "learning_rate": 4.238428981320975e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 38336176,
+      "step": 186
+    },
+    {
+      "epoch": 1.6530612244897958,
+      "grad_norm": 0.00755694042891264,
+      "learning_rate": 4.192666809072948e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 38548880,
+      "step": 187
+    },
+    {
+      "epoch": 1.6618863761720903,
+      "grad_norm": 0.01243317686021328,
+      "learning_rate": 4.146973971430333e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 38755920,
+      "step": 188
+    },
+    {
+      "epoch": 1.670711527854385,
+      "grad_norm": 0.006207725498825312,
+      "learning_rate": 4.101354392541061e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 38973328,
+      "step": 189
+    },
+    {
+      "epoch": 1.6795366795366795,
+      "grad_norm": 0.008532355539500713,
+      "learning_rate": 4.0558119902615174e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 39193232,
+      "step": 190
+    },
+    {
+      "epoch": 1.688361831218974,
+      "grad_norm": 0.008602111600339413,
+      "learning_rate": 4.010350675820091e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 39406608,
+      "step": 191
+    },
+    {
+      "epoch": 1.6971869829012687,
+      "grad_norm": 0.008903734385967255,
+      "learning_rate": 3.964974353481254e-05,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 39620160,
+      "step": 192
+    },
+    {
+      "epoch": 1.7060121345835633,
+      "grad_norm": 0.005871508736163378,
+      "learning_rate": 3.919686920210277e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 39815952,
+      "step": 193
+    },
+    {
+      "epoch": 1.7148372862658579,
+      "grad_norm": 0.008220325224101543,
+      "learning_rate": 3.874492265338544e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 40015408,
+      "step": 194
+    },
+    {
+      "epoch": 1.7236624379481522,
+      "grad_norm": 0.00940727163106203,
+      "learning_rate": 3.829394270229531e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 40215328,
+      "step": 195
+    },
+    {
+      "epoch": 1.7324875896304468,
+      "grad_norm": 0.005745697300881147,
+      "learning_rate": 3.784396807945477e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 40414384,
+      "step": 196
+    },
+    {
+      "epoch": 1.7413127413127412,
+      "grad_norm": 0.009524352848529816,
+      "learning_rate": 3.7395037429147615e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 40620656,
+      "step": 197
+    },
+    {
+      "epoch": 1.7501378929950357,
+      "grad_norm": 0.00809427909553051,
+      "learning_rate": 3.694718930600012e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 40847008,
+      "step": 198
+    },
+    {
+      "epoch": 1.7589630446773303,
+      "grad_norm": 0.0051635075360536575,
+      "learning_rate": 3.65004621716701e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 41036368,
+      "step": 199
+    },
+    {
+      "epoch": 1.767788196359625,
+      "grad_norm": 0.006504002492874861,
+      "learning_rate": 3.6054894391543646e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 41252976,
+      "step": 200
+    },
+    {
+      "epoch": 1.7766133480419195,
+      "grad_norm": 0.009855791926383972,
+      "learning_rate": 3.561052423144032e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 41465104,
+      "step": 201
+    },
+    {
+      "epoch": 1.785438499724214,
+      "grad_norm": 0.004304118454456329,
+      "learning_rate": 3.5167389854326905e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 41670800,
+      "step": 202
+    },
+    {
+      "epoch": 1.7942636514065087,
+      "grad_norm": 0.014682441018521786,
+      "learning_rate": 3.4725529317039754e-05,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 41883536,
+      "step": 203
+    },
+    {
+      "epoch": 1.803088803088803,
+      "grad_norm": 0.0061918287537992,
+      "learning_rate": 3.428498056701665e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 42083360,
+      "step": 204
+    },
+    {
+      "epoch": 1.8119139547710976,
+      "grad_norm": 0.009490927681326866,
+      "learning_rate": 3.38457814390377e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 42283120,
+      "step": 205
+    },
+    {
+      "epoch": 1.8207391064533922,
+      "grad_norm": 0.008434086106717587,
+      "learning_rate": 3.340796965197604e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 42499088,
+      "step": 206
+    },
+    {
+      "epoch": 1.8295642581356866,
+      "grad_norm": 0.004052174277603626,
+      "learning_rate": 3.297158280555862e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 42692976,
+      "step": 207
+    },
+    {
+      "epoch": 1.8383894098179812,
+      "grad_norm": 0.007411065977066755,
+      "learning_rate": 3.2536658377136935e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 42907216,
+      "step": 208
+    },
+    {
+      "epoch": 1.8472145615002757,
+      "grad_norm": 0.006996455602347851,
+      "learning_rate": 3.210323371846857e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 43112448,
+      "step": 209
+    },
+    {
+      "epoch": 1.8560397131825703,
+      "grad_norm": 0.006998082622885704,
+      "learning_rate": 3.167134605250938e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 43340096,
+      "step": 210
+    },
+    {
+      "epoch": 1.864864864864865,
+      "grad_norm": 0.006418649572879076,
+      "learning_rate": 3.124103247021657e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 43539664,
+      "step": 211
+    },
+    {
+      "epoch": 1.8736900165471595,
+      "grad_norm": 0.009151714853942394,
+      "learning_rate": 3.081232992736355e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 43727664,
+      "step": 212
+    },
+    {
+      "epoch": 1.882515168229454,
+      "grad_norm": 0.004692760296165943,
+      "learning_rate": 3.0385275241365962e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 43953584,
+      "step": 213
+    },
+    {
+      "epoch": 1.8913403199117484,
+      "grad_norm": 0.006455820985138416,
+      "learning_rate": 2.9959905088119776e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 44157504,
+      "step": 214
+    },
+    {
+      "epoch": 1.900165471594043,
+      "grad_norm": 0.006325691007077694,
+      "learning_rate": 2.9536255998851613e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 44350448,
+      "step": 215
+    },
+    {
+      "epoch": 1.9089906232763376,
+      "grad_norm": 0.006784004159271717,
+      "learning_rate": 2.9114364356981272e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 44561472,
+      "step": 216
+    },
+    {
+      "epoch": 1.917815774958632,
+      "grad_norm": 0.008874817751348019,
+      "learning_rate": 2.8694266394997238e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 44769936,
+      "step": 217
+    },
+    {
+      "epoch": 1.9266409266409266,
+      "grad_norm": 0.006964050233364105,
+      "learning_rate": 2.8275998191344888e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 44979344,
+      "step": 218
+    },
+    {
+      "epoch": 1.9354660783232212,
+      "grad_norm": 0.014264012686908245,
+      "learning_rate": 2.7859595667328026e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 45196944,
+      "step": 219
+    },
+    {
+      "epoch": 1.9442912300055157,
+      "grad_norm": 0.005279663018882275,
+      "learning_rate": 2.7445094584024067e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 45406832,
+      "step": 220
+    },
+    {
+      "epoch": 1.9531163816878103,
+      "grad_norm": 0.0171637125313282,
+      "learning_rate": 2.7032530539212658e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 45603120,
+      "step": 221
+    },
+    {
+      "epoch": 1.961941533370105,
+      "grad_norm": 0.007687513716518879,
+      "learning_rate": 2.6621938964318595e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 45805184,
+      "step": 222
+    },
+    {
+      "epoch": 1.9707666850523995,
+      "grad_norm": 0.0034611017908900976,
+      "learning_rate": 2.621335512136899e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 46001184,
+      "step": 223
+    },
+    {
+      "epoch": 1.9795918367346939,
+      "grad_norm": 0.004358428996056318,
+      "learning_rate": 2.5806814099964772e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 46206288,
+      "step": 224
+    },
+    {
+      "epoch": 1.9884169884169884,
+      "grad_norm": 0.008765267208218575,
+      "learning_rate": 2.540235081426736e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 46427344,
+      "step": 225
+    },
+    {
+      "epoch": 1.9972421400992828,
+      "grad_norm": 0.006889387033879757,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 46627344,
+      "step": 226
+    },
+    {
+      "epoch": 2.0088251516822946,
+      "grad_norm": 0.043494511395692825,
+      "learning_rate": 2.459979621146477e-05,
+      "loss": 0.0011,
+      "num_input_tokens_seen": 46901504,
+      "step": 227
+    },
+    {
+      "epoch": 2.017650303364589,
+      "grad_norm": 0.007718184031546116,
+      "learning_rate": 2.4201773818574956e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 47104400,
+      "step": 228
+    },
+    {
+      "epoch": 2.0264754550468838,
+      "grad_norm": 0.003912526648491621,
+      "learning_rate": 2.3805967003903333e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 47314176,
+      "step": 229
+    },
+    {
+      "epoch": 2.0353006067291783,
+      "grad_norm": 0.010783454403281212,
+      "learning_rate": 2.3412409759746528e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 47525264,
+      "step": 230
+    },
+    {
+      "epoch": 2.0441257584114725,
+      "grad_norm": 0.0026623259764164686,
+      "learning_rate": 2.302113588520578e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 47724528,
+      "step": 231
+    },
+    {
+      "epoch": 2.052950910093767,
+      "grad_norm": 0.00557671207934618,
+      "learning_rate": 2.2632178983284153e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 47932624,
+      "step": 232
+    },
+    {
+      "epoch": 2.0617760617760617,
+      "grad_norm": 0.003710981458425522,
+      "learning_rate": 2.2245572458000712e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 48148608,
+      "step": 233
+    },
+    {
+      "epoch": 2.0706012134583562,
+      "grad_norm": 0.009742701426148415,
+      "learning_rate": 2.1861349511521815e-05,
+      "loss": 0.0025,
+      "num_input_tokens_seen": 48373632,
+      "step": 234
+    },
+    {
+      "epoch": 2.079426365140651,
+      "grad_norm": 0.009755464270710945,
+      "learning_rate": 2.147954314130955e-05,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 48586512,
+      "step": 235
+    },
+    {
+      "epoch": 2.0882515168229454,
+      "grad_norm": 0.002706202445551753,
+      "learning_rate": 2.1100186137288e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 48793568,
+      "step": 236
+    },
+    {
+      "epoch": 2.09707666850524,
+      "grad_norm": 0.005180325359106064,
+      "learning_rate": 2.072331107902713e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 49006224,
+      "step": 237
+    },
+    {
+      "epoch": 2.1059018201875346,
+      "grad_norm": 0.005968959536403418,
+      "learning_rate": 2.0348950332944834e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 49217632,
+      "step": 238
+    },
+    {
+      "epoch": 2.114726971869829,
+      "grad_norm": 0.0063306307420134544,
+      "learning_rate": 1.9977136049527345e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 49426624,
+      "step": 239
+    },
+    {
+      "epoch": 2.1235521235521237,
+      "grad_norm": 0.005157762672752142,
+      "learning_rate": 1.960790016056801e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 49623376,
+      "step": 240
+    },
+    {
+      "epoch": 2.132377275234418,
+      "grad_norm": 0.005218483041971922,
+      "learning_rate": 1.9241274376425e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 49828144,
+      "step": 241
+    },
+    {
+      "epoch": 2.1412024269167125,
+      "grad_norm": 0.00744604179635644,
+      "learning_rate": 1.8877290183298057e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 50018448,
+      "step": 242
+    },
+    {
+      "epoch": 2.150027578599007,
+      "grad_norm": 0.005399591755121946,
+      "learning_rate": 1.8515978840524302e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 50218176,
+      "step": 243
+    },
+    {
+      "epoch": 2.1588527302813016,
+      "grad_norm": 0.005761398002505302,
+      "learning_rate": 1.815737137789377e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 50424896,
+      "step": 244
+    },
+    {
+      "epoch": 2.1676778819635962,
+      "grad_norm": 0.006964447908103466,
+      "learning_rate": 1.7801498592984446e-05,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 50635088,
+      "step": 245
+    },
+    {
+      "epoch": 2.176503033645891,
+      "grad_norm": 0.002962745726108551,
+      "learning_rate": 1.7448391048517376e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 50849552,
+      "step": 246
+    },
+    {
+      "epoch": 2.1853281853281854,
+      "grad_norm": 0.005332667380571365,
+      "learning_rate": 1.7098079069731958e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 51037776,
+      "step": 247
+    },
+    {
+      "epoch": 2.19415333701048,
+      "grad_norm": 0.006928949151188135,
+      "learning_rate": 1.6750592741781497e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 51242672,
+      "step": 248
+    },
+    {
+      "epoch": 2.2029784886927746,
+      "grad_norm": 0.004213888198137283,
+      "learning_rate": 1.640596190714947e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 51437008,
+      "step": 249
+    },
+    {
+      "epoch": 2.211803640375069,
+      "grad_norm": 0.010446918196976185,
+      "learning_rate": 1.6064216163086716e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 51641264,
+      "step": 250
+    },
+    {
+      "epoch": 2.2206287920573633,
+      "grad_norm": 0.004029524512588978,
+      "learning_rate": 1.5725384859069455e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 51842592,
+      "step": 251
+    },
+    {
+      "epoch": 2.229453943739658,
+      "grad_norm": 0.006790219806134701,
+      "learning_rate": 1.538949709427886e-05,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 52047456,
+      "step": 252
+    },
+    {
+      "epoch": 2.2382790954219525,
+      "grad_norm": 0.003987099044024944,
+      "learning_rate": 1.5056581715101886e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 52242208,
+      "step": 253
+    },
+    {
+      "epoch": 2.247104247104247,
+      "grad_norm": 0.008930574171245098,
+      "learning_rate": 1.472666731265394e-05,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 52436800,
+      "step": 254
+    },
+    {
+      "epoch": 2.2559293987865416,
+      "grad_norm": 0.004108684603124857,
+      "learning_rate": 1.4399782220323515e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 52624752,
+      "step": 255
+    },
+    {
+      "epoch": 2.2647545504688362,
+      "grad_norm": 0.00732703972607851,
+      "learning_rate": 1.4075954511338785e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 52836384,
+      "step": 256
+    },
+    {
+      "epoch": 2.273579702151131,
+      "grad_norm": 0.006608397234231234,
+      "learning_rate": 1.3755211996356687e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 53059296,
+      "step": 257
+    },
+    {
+      "epoch": 2.2824048538334254,
+      "grad_norm": 0.002376733347773552,
+      "learning_rate": 1.3437582221074573e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 53267440,
+      "step": 258
+    },
+    {
+      "epoch": 2.29123000551572,
+      "grad_norm": 0.004921163432300091,
+      "learning_rate": 1.3123092463864456e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 53501008,
+      "step": 259
+    },
+    {
+      "epoch": 2.300055157198014,
+      "grad_norm": 0.0034377635456621647,
+      "learning_rate": 1.2811769733430406e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 53700432,
+      "step": 260
+    },
+    {
+      "epoch": 2.3088803088803087,
+      "grad_norm": 0.006821690127253532,
+      "learning_rate": 1.250364076648894e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 53919616,
+      "step": 261
+    },
+    {
+      "epoch": 2.3177054605626033,
+      "grad_norm": 0.004776927176862955,
+      "learning_rate": 1.2198732025472876e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 54130528,
+      "step": 262
+    },
+    {
+      "epoch": 2.326530612244898,
+      "grad_norm": 0.004824692849069834,
+      "learning_rate": 1.1897069696258755e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 54350560,
+      "step": 263
+    },
+    {
+      "epoch": 2.3353557639271925,
+      "grad_norm": 0.005174586083739996,
+      "learning_rate": 1.1598679685917901e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 54542224,
+      "step": 264
+    },
+    {
+      "epoch": 2.344180915609487,
+      "grad_norm": 0.012352543883025646,
+      "learning_rate": 1.1303587620491513e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 54745136,
+      "step": 265
+    },
+    {
+      "epoch": 2.3530060672917816,
+      "grad_norm": 0.005056153051555157,
+      "learning_rate": 1.1011818842789928e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 54957584,
+      "step": 266
+    },
+    {
+      "epoch": 2.361831218974076,
+      "grad_norm": 0.010525842197239399,
+      "learning_rate": 1.0723398410216084e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 55162496,
+      "step": 267
+    },
+    {
+      "epoch": 2.370656370656371,
+      "grad_norm": 0.0092442212626338,
+      "learning_rate": 1.0438351092613569e-05,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 55376544,
+      "step": 268
+    },
+    {
+      "epoch": 2.3794815223386654,
+      "grad_norm": 0.00699999462813139,
+      "learning_rate": 1.0156701370139454e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 55583072,
+      "step": 269
+    },
+    {
+      "epoch": 2.38830667402096,
+      "grad_norm": 0.007677710149437189,
+      "learning_rate": 9.878473431161767e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 55801200,
+      "step": 270
+    },
+    {
+      "epoch": 2.397131825703254,
+      "grad_norm": 0.003174175275489688,
+      "learning_rate": 9.603691170182317e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 55998080,
+      "step": 271
+    },
+    {
+      "epoch": 2.4059569773855487,
+      "grad_norm": 0.005871200002729893,
+      "learning_rate": 9.33237818578449e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 56200448,
+      "step": 272
+    },
+    {
+      "epoch": 2.4147821290678433,
+      "grad_norm": 0.00371691957116127,
+      "learning_rate": 9.064557778606631e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 56400416,
+      "step": 273
+    },
+    {
+      "epoch": 2.423607280750138,
+      "grad_norm": 0.007599337492138147,
+      "learning_rate": 8.800252949340998e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 56606128,
+      "step": 274
+    },
+    {
+      "epoch": 2.4324324324324325,
+      "grad_norm": 0.0015243644593283534,
+      "learning_rate": 8.539486396758356e-06,
+      "loss": 0.0,
+      "num_input_tokens_seen": 56797824,
+      "step": 275
+    },
+    {
+      "epoch": 2.441257584114727,
+      "grad_norm": 0.0030196798034012318,
+      "learning_rate": 8.28228051575864e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 57006384,
+      "step": 276
+    },
+    {
+      "epoch": 2.4500827357970216,
+      "grad_norm": 0.005347589962184429,
+      "learning_rate": 8.02865739544767e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 57207824,
+      "step": 277
+    },
+    {
+      "epoch": 2.458907887479316,
+      "grad_norm": 0.005150883924216032,
+      "learning_rate": 7.778638817240042e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 57415152,
+      "step": 278
+    },
+    {
+      "epoch": 2.467733039161611,
+      "grad_norm": 0.006857512053102255,
+      "learning_rate": 7.532246252988617e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 57628096,
+      "step": 279
+    },
+    {
+      "epoch": 2.476558190843905,
+      "grad_norm": 0.005364645272493362,
+      "learning_rate": 7.289500863140414e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 57824064,
+      "step": 280
+    },
+    {
+      "epoch": 2.4853833425261995,
+      "grad_norm": 0.007198365870863199,
+      "learning_rate": 7.05042349491935e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 58042720,
+      "step": 281
+    },
+    {
+      "epoch": 2.494208494208494,
+      "grad_norm": 0.005014900583773851,
+      "learning_rate": 6.815034680535915e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 58255408,
+      "step": 282
+    },
+    {
+      "epoch": 2.5030336458907887,
+      "grad_norm": 0.008873779326677322,
+      "learning_rate": 6.5833546354237556e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 58464800,
+      "step": 283
+    },
+    {
+      "epoch": 2.5118587975730833,
+      "grad_norm": 0.0044725253246724606,
+      "learning_rate": 6.355403256503595e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 58672496,
+      "step": 284
+    },
+    {
+      "epoch": 2.520683949255378,
+      "grad_norm": 0.0047348616644740105,
+      "learning_rate": 6.1312001204745115e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 58898256,
+      "step": 285
+    },
+    {
+      "epoch": 2.5295091009376725,
+      "grad_norm": 0.00710884016007185,
+      "learning_rate": 5.910764482132575e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 59107152,
+      "step": 286
+    },
+    {
+      "epoch": 2.538334252619967,
+      "grad_norm": 0.007686229422688484,
+      "learning_rate": 5.6941152727173265e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 59307664,
+      "step": 287
+    },
+    {
+      "epoch": 2.5471594043022616,
+      "grad_norm": 0.014555118046700954,
+      "learning_rate": 5.481271098285817e-06,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 59514736,
+      "step": 288
+    },
+    {
+      "epoch": 2.5559845559845558,
+      "grad_norm": 0.0028200196102261543,
+      "learning_rate": 5.272250238114856e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 59712512,
+      "step": 289
+    },
+    {
+      "epoch": 2.564809707666851,
+      "grad_norm": 0.004194322973489761,
+      "learning_rate": 5.067070643131055e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 59910000,
+      "step": 290
+    },
+    {
+      "epoch": 2.573634859349145,
+      "grad_norm": 0.006987538188695908,
+      "learning_rate": 4.865749934369223e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 60116400,
+      "step": 291
+    },
+    {
+      "epoch": 2.5824600110314395,
+      "grad_norm": 0.003778768004849553,
+      "learning_rate": 4.668305401459022e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 60320368,
+      "step": 292
+    },
+    {
+      "epoch": 2.591285162713734,
+      "grad_norm": 0.003472360782325268,
+      "learning_rate": 4.474754001140191e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 60536528,
+      "step": 293
+    },
+    {
+      "epoch": 2.6001103143960287,
+      "grad_norm": 0.009052475914359093,
+      "learning_rate": 4.285112355806192e-06,
+      "loss": 0.001,
+      "num_input_tokens_seen": 60743120,
+      "step": 294
+    },
+    {
+      "epoch": 2.6089354660783233,
+      "grad_norm": 0.0060082292184233665,
+      "learning_rate": 4.099396752076745e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 60942704,
+      "step": 295
+    },
+    {
+      "epoch": 2.617760617760618,
+      "grad_norm": 0.0075798071920871735,
+      "learning_rate": 3.917623139399018e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 61140128,
+      "step": 296
+    },
+    {
+      "epoch": 2.6265857694429124,
+      "grad_norm": 0.0055752964690327644,
+      "learning_rate": 3.7398071286779857e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 61334224,
+      "step": 297
+    },
+    {
+      "epoch": 2.6354109211252066,
+      "grad_norm": 0.007863204926252365,
+      "learning_rate": 3.5659639909356723e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 61543280,
+      "step": 298
+    },
+    {
+      "epoch": 2.6442360728075016,
+      "grad_norm": 0.006538075394928455,
+      "learning_rate": 3.3961086559996803e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 61750720,
+      "step": 299
+    },
+    {
+      "epoch": 2.6530612244897958,
+      "grad_norm": 0.002779777627438307,
+      "learning_rate": 3.230255711220992e-06,
+      "loss": 0.0,
+      "num_input_tokens_seen": 61945952,
+      "step": 300
+    },
+    {
+      "epoch": 2.6618863761720903,
+      "grad_norm": 0.004271807614713907,
+      "learning_rate": 3.0684194002212287e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 62155632,
+      "step": 301
+    },
+    {
+      "epoch": 2.670711527854385,
+      "grad_norm": 0.00638817623257637,
+      "learning_rate": 2.910613621669356e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 62353216,
+      "step": 302
+    },
+    {
+      "epoch": 2.6795366795366795,
+      "grad_norm": 0.00442032516002655,
+      "learning_rate": 2.7568519280880558e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 62544128,
+      "step": 303
+    },
+    {
+      "epoch": 2.688361831218974,
+      "grad_norm": 0.008686737157404423,
+      "learning_rate": 2.607147524689829e-06,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 62752688,
+      "step": 304
+    },
+    {
+      "epoch": 2.6971869829012687,
+      "grad_norm": 0.0059651597402989864,
+      "learning_rate": 2.4615132682429374e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 62963296,
+      "step": 305
+    },
+    {
+      "epoch": 2.7060121345835633,
+      "grad_norm": 0.0056177834048867226,
+      "learning_rate": 2.3199616659672354e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 63161904,
+      "step": 306
+    },
+    {
+      "epoch": 2.714837286265858,
+      "grad_norm": 0.0029979923274368048,
+      "learning_rate": 2.182504874460006e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 63365744,
+      "step": 307
+    },
+    {
+      "epoch": 2.7236624379481524,
+      "grad_norm": 0.004314000252634287,
+      "learning_rate": 2.049154698651989e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 63571808,
+      "step": 308
+    },
+    {
+      "epoch": 2.7324875896304466,
+      "grad_norm": 0.006837273947894573,
+      "learning_rate": 1.919922590793549e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 63768960,
+      "step": 309
+    },
+    {
+      "epoch": 2.741312741312741,
+      "grad_norm": 0.0037646403070539236,
+      "learning_rate": 1.7948196494711188e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 63979648,
+      "step": 310
+    },
+    {
+      "epoch": 2.7501378929950357,
+      "grad_norm": 0.0031723175197839737,
+      "learning_rate": 1.6738566186540627e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 64189712,
+      "step": 311
+    },
+    {
+      "epoch": 2.7589630446773303,
+      "grad_norm": 0.005477920174598694,
+      "learning_rate": 1.5570438867719694e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 64400624,
+      "step": 312
+    },
+    {
+      "epoch": 2.767788196359625,
+      "grad_norm": 0.006315939594060183,
+      "learning_rate": 1.4443914858224938e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 64626320,
+      "step": 313
+    },
+    {
+      "epoch": 2.7766133480419195,
+      "grad_norm": 0.004753002431243658,
+      "learning_rate": 1.3359090905097848e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 64826480,
+      "step": 314
+    },
+    {
+      "epoch": 2.785438499724214,
+      "grad_norm": 0.00812880601733923,
+      "learning_rate": 1.2316060174136002e-06,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 65031984,
+      "step": 315
+    },
+    {
+      "epoch": 2.7942636514065087,
+      "grad_norm": 0.0029212606605142355,
+      "learning_rate": 1.1314912241892183e-06,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 65239456,
+      "step": 316
+    },
+    {
+      "epoch": 2.8030888030888033,
+      "grad_norm": 0.006850802339613438,
+      "learning_rate": 1.0355733087981378e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 65433888,
+      "step": 317
+    },
+    {
+      "epoch": 2.8119139547710974,
+      "grad_norm": 0.0020711093675345182,
+      "learning_rate": 9.43860508769645e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 65638288,
+      "step": 318
+    },
+    {
+      "epoch": 2.8207391064533924,
+      "grad_norm": 0.004868640564382076,
+      "learning_rate": 8.563607004934193e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 65855952,
+      "step": 319
+    },
+    {
+      "epoch": 2.8295642581356866,
+      "grad_norm": 0.006297328509390354,
+      "learning_rate": 7.730813985430407e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 66070192,
+      "step": 320
+    },
+    {
+      "epoch": 2.838389409817981,
+      "grad_norm": 0.0036759376525878906,
+      "learning_rate": 6.940297550306896e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 66283808,
+      "step": 321
+    },
+    {
+      "epoch": 2.8472145615002757,
+      "grad_norm": 0.0120092136785388,
+      "learning_rate": 6.192125589928821e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 66507776,
+      "step": 322
+    },
+    {
+      "epoch": 2.8560397131825703,
+      "grad_norm": 0.005414010491222143,
+      "learning_rate": 5.486362358074094e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 66708320,
+      "step": 323
+    },
+    {
+      "epoch": 2.864864864864865,
+      "grad_norm": 0.007992051541805267,
+      "learning_rate": 4.823068466415615e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 66910032,
+      "step": 324
+    },
+    {
+      "epoch": 2.8736900165471595,
+      "grad_norm": 0.006493248511105776,
+      "learning_rate": 4.202300879315446e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 67112784,
+      "step": 325
+    },
+    {
+      "epoch": 2.882515168229454,
+      "grad_norm": 0.004381334874778986,
+      "learning_rate": 3.624112908932942e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 67306464,
+      "step": 326
+    },
+    {
+      "epoch": 2.8913403199117482,
+      "grad_norm": 0.00577085604891181,
+      "learning_rate": 3.088554210646133e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 67504720,
+      "step": 327
+    },
+    {
+      "epoch": 2.9001654715940433,
+      "grad_norm": 0.003793071024119854,
+      "learning_rate": 2.595670778787196e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 67694048,
+      "step": 328
+    },
+    {
+      "epoch": 2.9089906232763374,
+      "grad_norm": 0.00835067592561245,
+      "learning_rate": 2.1455049426926666e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 67895008,
+      "step": 329
+    },
+    {
+      "epoch": 2.917815774958632,
+      "grad_norm": 0.005372443702071905,
+      "learning_rate": 1.7380953630678488e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 68093168,
+      "step": 330
+    },
+    {
+      "epoch": 2.9266409266409266,
+      "grad_norm": 0.010219305753707886,
+      "learning_rate": 1.373477028666803e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 68305568,
+      "step": 331
+    },
+    {
+      "epoch": 2.935466078323221,
+      "grad_norm": 0.0038206197787076235,
+      "learning_rate": 1.0516812532873621e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 68506384,
+      "step": 332
+    },
+    {
+      "epoch": 2.9442912300055157,
+      "grad_norm": 0.007432411424815655,
+      "learning_rate": 7.727356730820035e-08,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 68716160,
+      "step": 333
+    },
+    {
+      "epoch": 2.9531163816878103,
+      "grad_norm": 0.004036502446979284,
+      "learning_rate": 5.3666424418413744e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 68918048,
+      "step": 334
+    },
+    {
+      "epoch": 2.961941533370105,
+      "grad_norm": 0.0045955548994243145,
+      "learning_rate": 3.4348724065119685e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 69129152,
+      "step": 335
+    },
+    {
+      "epoch": 2.9707666850523995,
+      "grad_norm": 0.012164157815277576,
+      "learning_rate": 1.9322125272297488e-08,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 69328576,
+      "step": 336
+    },
+    {
+      "epoch": 2.979591836734694,
+      "grad_norm": 0.0029640356078743935,
+      "learning_rate": 8.587918539726402e-09,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 69537232,
+      "step": 337
+    },
+    {
+      "epoch": 2.988416988416988,
+      "grad_norm": 0.005239939782768488,
+      "learning_rate": 2.1470257321298813e-09,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 69761008,
+      "step": 338
+    },
+    {
+      "epoch": 2.997242140099283,
+      "grad_norm": 0.0060053626075387,
+      "learning_rate": 0.0,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 69953200,
+      "step": 339
+    },
+    {
+      "epoch": 2.997242140099283,
+      "num_input_tokens_seen": 69953200,
+      "step": 339,
+      "total_flos": 2.976146663409713e+18,
+      "train_loss": 0.004280612113766934,
+      "train_runtime": 8852.4475,
+      "train_samples_per_second": 4.914,
+      "train_steps_per_second": 0.038
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 339,
+  "num_input_tokens_seen": 69953200,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.976146663409713e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96fd9d2ed84b9224ea0980ce9d54347d209b8b7db550f9e136d47533c595945b
+size 5688

plan/sft-qwen2.5-7b-instruct-graph-planning-bs128/training_args.yaml

@@ -0,0 +1,33 @@
+bf16: true
+cutoff_len: 4096
+dataset: graph_planning_train
+dataset_dir: data
+ddp_timeout: 180000000
+do_train: true
+finetuning_type: lora
+flash_attn: auto
+gradient_accumulation_steps: 16
+include_num_input_tokens_seen: true
+learning_rate: 0.0001
+logging_steps: 1
+lora_alpha: 16
+lora_dropout: 0
+lora_rank: 8
+lora_target: all
+lr_scheduler_type: cosine
+max_grad_norm: 1.0
+max_samples: 100000
+model_name_or_path: /nas/shared/ma4agi/model/Qwen2.5-7B-Instruct
+num_train_epochs: 3.0
+optim: adamw_torch
+output_dir: saves/Qwen2.5-7B-Instruct/lora/sft-qwen2.5-7b-instruct-graph-planning-bs128
+packing: false
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+report_to: none
+save_steps: 100
+stage: sft
+template: qwen
+trust_remote_code: true
+warmup_steps: 0