Model save

README.md

@@ -1,10 +1,9 @@
 ---
 library_name: transformers
-license: llama3.1
-base_model: meta-llama/Meta-Llama-3.1-8B
+license: apache-2.0
+base_model: Qwen/Qwen2.5-7B-Instruct
 tags:
 - llama-factory
-- full
 - generated_from_trainer
 model-index:
 - name: llama3-1_8b_4o_annotated_olympiads
@@ -16,9 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-1_8b_4o_annotated_olympiads
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/4o_annotated_olympiads dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.7641
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on an unknown dataset.
 
 ## Model description
 
@@ -37,25 +34,22 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-06
-- train_batch_size: 16
+- learning_rate: 1e-05
+- train_batch_size: 1
 - eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 32
-- total_train_batch_size: 512
+- gradient_accumulation_steps: 3
+- total_train_batch_size: 96
 - total_eval_batch_size: 256
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: constant
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
 - num_epochs: 3.0
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 0.8464        | 1.0   | 25   | 0.8046          |
-| 0.7628        | 2.0   | 50   | 0.7725          |
-| 0.7237        | 3.0   | 75   | 0.7641          |
 
 
 ### Framework versions

generation_config.json

@@ -1,9 +1,14 @@
 {
-  "_from_model_config": true,
-  "bos_token_id": 128000,
+  "bos_token_id": 151643,
   "do_sample": true,
-  "eos_token_id": 128001,
-  "temperature": 0.6,
-  "top_p": 0.9,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
   "transformers_version": "4.46.1"
 }