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/c0d8bbe8a93c1465_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 3494cd1e-059c-4d32-8d9e-81f4630dfa3e
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 3494cd1e-059c-4d32-8d9e-81f4630dfa3e
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 3494cd1e-059c-4d32-8d9e-81f4630dfa3e
 
 This model is a fine-tuned version of [Qwen/Qwen2-0.5B](https://huggingface.co/Qwen/Qwen2-0.5B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.3915
 
 ## 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        | 0.0027 | 1    | 1.4483          |
+| 1.5137        | 0.0265 | 10   | 1.4359          |
+| 1.2843        | 0.0531 | 20   | 1.4107          |
+| 1.3807        | 0.0796 | 30   | 1.3976          |
+| 1.4292        | 0.1062 | 40   | 1.3924          |
+| 1.3027        | 0.1327 | 50   | 1.3915          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2bec3b583f4c6a7ea973177398ded3b91dfab8d658264bb3e14f533d7733988a
+oid sha256:02d10a45706862d3659db80d957101ef80ef21e7ef0cf58627b08e0a0589ff0d
 size 17717130