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/6beec10d32177767_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 11a1d9f6-78ee-419c-a6c5-ed04e332eef8
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 11a1d9f6-78ee-419c-a6c5-ed04e332eef8
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 6459948a-3de2-56e8-db03-fcb4c6c06c9c
 
 This model is a fine-tuned version of [sethuiyer/Medichat-Llama3-8B](https://huggingface.co/sethuiyer/Medichat-Llama3-8B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.1709
 
 ## 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        | 0.0043 | 1    | 5.1333          |
+| 4.5264        | 0.0426 | 10   | 4.5604          |
+| 3.6324        | 0.0851 | 20   | 3.4525          |
+| 3.1833        | 0.1277 | 30   | 3.2465          |
+| 3.2878        | 0.1702 | 40   | 3.1812          |
+| 3.1203        | 0.2128 | 50   | 3.1709          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6aee51317acddb6240532dd93e9102a0544696ac0cbeb100c9000c8731a31e57
+oid sha256:affa37780a0deb58dfdef93b98126ad0545cc33ff431547f7822a7de12e68d6d
 size 84047370