End of training

README.md

@@ -66,7 +66,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 50
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/197929514d450a21_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 9938f538-fc0f-4031-9057-5ff481ca4356
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 9938f538-fc0f-4031-9057-5ff481ca4356
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 7abb3d7a-f851-4349-a78d-18d89a3841b6
 
 This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.6555
 
 ## Model description
 
@@ -128,14 +130,19 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 8
 - 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
-- lr_scheduler_warmup_steps: 2
-- training_steps: 1
+- lr_scheduler_warmup_steps: 10
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0040 | 1    | 2.9856          |
+| 2.8445        | 0.0399 | 10   | 2.8598          |
+| 2.2853        | 0.0798 | 20   | 2.2561          |
+| 1.9431        | 0.1198 | 30   | 1.8083          |
+| 1.5311        | 0.1597 | 40   | 1.6713          |
+| 1.565         | 0.1996 | 50   | 1.6555          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4d55b0aa8b7c6c8914cbbc7517c674b194e175d1da0d9af6e196b2869888933
+oid sha256:177e7d256940f5e4d64b308fbda36d1ee7d195745b9c0f07de54584e19599ce5
 size 37070634