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/0c24410fedb9faaf_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: ef415b9d-7bd0-4d4b-9c00-9205daa4b4a8
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: ef415b9d-7bd0-4d4b-9c00-9205daa4b4a8
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # ef415b9d-7bd0-4d4b-9c00-9205daa4b4a8
 
 This model is a fine-tuned version of [DeepMount00/Llama-3-8b-Ita](https://huggingface.co/DeepMount00/Llama-3-8b-Ita) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4999
 
 ## Model description
 
@@ -129,14 +131,19 @@ 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        | 0.0001 | 1    | 0.6391          |
+| 0.601         | 0.0014 | 10   | 0.6055          |
+| 0.5578        | 0.0029 | 20   | 0.5375          |
+| 0.5012        | 0.0043 | 30   | 0.5091          |
+| 0.4668        | 0.0057 | 40   | 0.5011          |
+| 0.4966        | 0.0072 | 50   | 0.4999          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df5d2273ded02f9bd1382d1abf59f026167226e0652305c544c49ee864c05631
+oid sha256:c3ea1773c201179962173f1ee94410a44cae77828b82810f4eb08a9382ec5156
 size 84047370