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: 50
+max_steps: 1
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/9ccbbf70abb8429b_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 6086b8cf-e0ed-479e-81fa-2b15c16e4c40
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 6086b8cf-e0ed-479e-81fa-2b15c16e4c40
-warmup_steps: 10
+warmup_steps: 1
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,8 +101,6 @@ xformers_attention: null
 # 6086b8cf-e0ed-479e-81fa-2b15c16e4c40
 
 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: 0.2740
 
 ## Model description
 
@@ -129,19 +127,14 @@ 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: 10
-- training_steps: 50
+- lr_scheduler_warmup_steps: 2
+- training_steps: 1
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0003 | 1    | 12.1248         |
-| 10.3443       | 0.0034 | 10   | 8.4496          |
-| 1.7248        | 0.0068 | 20   | 0.8425          |
-| 0.4068        | 0.0101 | 30   | 0.3390          |
-| 0.1524        | 0.0135 | 40   | 0.2932          |
-| 0.1067        | 0.0169 | 50   | 0.2740          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61d08536dc37468a0aa93f36c588a5eb8ce00d124a336b1d7b2417219976f68d
+oid sha256:d1a86380213c693a1dd884faeedb4e4c1e2fc13b4a7b3b8b39680b86232fca1d
 size 48768810