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/611bd1a7e6e7f93c_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: a7a5d03b-19bc-4ae1-ba3c-e7e47a68ecb3
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: a7a5d03b-19bc-4ae1-ba3c-e7e47a68ecb3
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # 039a98cf-b040-eae2-bfb7-56599e8c8850
 
 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: 1.5716
 
 ## Model description
 
@@ -130,14 +132,19 @@ 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        | 0.0012 | 1    | 2.0840          |
+| 1.892         | 0.0121 | 10   | 1.8972          |
+| 1.7319        | 0.0241 | 20   | 1.6522          |
+| 1.5354        | 0.0362 | 30   | 1.5947          |
+| 1.539         | 0.0483 | 40   | 1.5754          |
+| 1.5315        | 0.0603 | 50   | 1.5716          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:049c7421f07a76d91bf64d7f3699eba9a2f2552eb1b7097d5aefe9267d0964b9
+oid sha256:fd77c9abb759860a4fb6a112c3d1c05cbed52f0c012748dd54736921818a617b
 size 84047370