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/e4f277da329e1205_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 5dc1f873-17e9-430a-8542-f91444d001f6
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 5dc1f873-17e9-430a-8542-f91444d001f6
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 18885749-2954-4609-9c66-079d7a6e182d
 
 This model is a fine-tuned version of [unsloth/Llama-3.1-Storm-8B](https://huggingface.co/unsloth/Llama-3.1-Storm-8B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8972
 
 ## 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        | 0.0002 | 1    | 2.2553          |
+| 1.9819        | 0.0015 | 10   | 1.7232          |
+| 1.103         | 0.0031 | 20   | 1.0012          |
+| 0.9432        | 0.0046 | 30   | 0.9205          |
+| 0.9569        | 0.0061 | 40   | 0.9023          |
+| 0.885         | 0.0077 | 50   | 0.8972          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9c48c04c65b8cb09cbf3f01cfde3e032eb9b2765a7aaee2ab0c48f329f1e22a
+oid sha256:4ead1b09a6d90db17be160210ac80e2ce13128b72444eb640f0a55e41d744d51
 size 84047370