End of training

README.md

@@ -65,7 +65,7 @@ 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/f50886842ccdb2ca_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 49a896dc-dc24-4334-b960-a4260210a4ec
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 49a896dc-dc24-4334-b960-a4260210a4ec
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 3faf3ee1-6553-46ea-b363-bca9d4089d40
 
 This model is a fine-tuned version of [unsloth/Llama-3.2-3B](https://huggingface.co/unsloth/Llama-3.2-3B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.7986
 
 ## Model description
 
@@ -127,14 +129,19 @@ 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        | 0.0020 | 1    | 2.3531          |
+| 2.1891        | 0.0196 | 10   | 2.1915          |
+| 1.799         | 0.0393 | 20   | 1.8569          |
+| 1.7006        | 0.0589 | 30   | 1.8206          |
+| 1.9341        | 0.0786 | 40   | 1.8029          |
+| 1.8628        | 0.0982 | 50   | 1.7986          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b533b0e2407ca5b14cafc9bc26b7eee33b89a0a8838a3f4227cc2872a98879d
+oid sha256:9f1664c6bad84c0a54e66549baab4b0934c7c6c8902997837c13fe8513426a66
 size 48768810