End of training

README.md

@@ -1,11 +1,11 @@
 ---
 library_name: transformers
-base_model: fxmarty/tiny-llama-fast-tokenizer
+base_model: katuni4ka/tiny-random-qwen1.5-moe
 tags:
 - axolotl
 - generated_from_trainer
 model-index:
-- name: b1c9c4ec-ffa2-429d-9c5b-90b5979c502d
+- name: e61e89f0-854a-4922-8d25-dae435e91af0
   results: []
 ---
 
@@ -17,20 +17,21 @@ should probably proofread and complete it, then remove this comment. -->
 
 axolotl version: `0.4.1`
 ```yaml
-base_model: fxmarty/tiny-llama-fast-tokenizer
+base_model: katuni4ka/tiny-random-qwen1.5-moe
 batch_size: 32
 bf16: true
 chat_template: tokenizer_default_fallback_alpaca
 datasets:
 - data_files:
-  - fc6136aac03f618a_train_data.json
+  - 95544452e61c7393_train_data.json
   ds_type: json
   format: custom
-  path: /workspace/input_data/fc6136aac03f618a_train_data.json
+  path: /workspace/input_data/95544452e61c7393_train_data.json
   type:
-    field_instruction: text
-    field_output: title
-    format: '{instruction}'
+    field_input: input
+    field_instruction: instruction
+    field_output: output
+    format: '{instruction} {input}'
     no_input_format: '{instruction}'
     system_format: '{system}'
     system_prompt: ''
@@ -39,7 +40,7 @@ flash_attention: true
 gpu_memory_limit: 80GiB
 gradient_checkpointing: true
 group_by_length: true
-hub_model_id: willtensora/b1c9c4ec-ffa2-429d-9c5b-90b5979c502d
+hub_model_id: willtensora/e61e89f0-854a-4922-8d25-dae435e91af0
 hub_strategy: checkpoint
 learning_rate: 0.0002
 logging_steps: 10
@@ -55,15 +56,13 @@ sample_packing: false
 save_steps: 40
 save_total_limit: 1
 sequence_len: 2048
-special_tokens:
-  pad_token: </s>
-tokenizer_type: LlamaTokenizerFast
+tokenizer_type: Qwen2TokenizerFast
 train_on_inputs: false
 trust_remote_code: true
 val_set_size: 0.1
 wandb_entity: ''
 wandb_mode: online
-wandb_name: fxmarty/tiny-llama-fast-tokenizer-/workspace/input_data/fc6136aac03f618a_train_data.json
+wandb_name: katuni4ka/tiny-random-qwen1.5-moe-/workspace/input_data/95544452e61c7393_train_data.json
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: default
@@ -74,9 +73,11 @@ xformers_attention: true
 
 </details><br>
 
-# b1c9c4ec-ffa2-429d-9c5b-90b5979c502d
+# e61e89f0-854a-4922-8d25-dae435e91af0
 
-This model is a fine-tuned version of [fxmarty/tiny-llama-fast-tokenizer](https://huggingface.co/fxmarty/tiny-llama-fast-tokenizer) on the None dataset.
+This model is a fine-tuned version of [katuni4ka/tiny-random-qwen1.5-moe](https://huggingface.co/katuni4ka/tiny-random-qwen1.5-moe) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.6281
 
 ## Model description
 
@@ -105,13 +106,16 @@ The following hyperparameters were used during training:
 - total_eval_batch_size: 32
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
-- training_steps: 18
+- lr_scheduler_warmup_steps: 2
+- training_steps: 40
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.0071 | 1    | 10.3739         |
+| No log        | 0.0031 | 1    | 11.9223         |
+| 11.7325       | 0.0629 | 20   | 11.6783         |
+| 11.6304       | 0.1258 | 40   | 11.6281         |
 
 
 ### Framework versions

generation_config.json

@@ -1,8 +1,7 @@
 {
   "_from_model_config": true,
-  "bos_token_id": 0,
+  "bos_token_id": 151643,
   "do_sample": true,
-  "eos_token_id": 1,
-  "pad_token_id": 1,
+  "eos_token_id": 151643,
   "transformers_version": "4.46.0"
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ecbabedee28483af8dce99f4dd8fe36ef9c6c66877e669db930fe3569128330
-size 2071661
+oid sha256:244c30fce0d5c4892e3b25d25e50c952fa49cb08493bb32684f850179545a7e3
+size 19817334