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/37077c82b6e4beed_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: 088e9221-610f-483f-b2eb-4e1cee0c9f08
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 088e9221-610f-483f-b2eb-4e1cee0c9f08
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # 74b1b5ed-9ac1-2600-7406-2ece8c173124
 
 This model is a fine-tuned version of [heegyu/WizardVicuna-open-llama-3b-v2](https://huggingface.co/heegyu/WizardVicuna-open-llama-3b-v2) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3039
 
 ## 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        | 0.0000 | 1    | 0.3857          |
+| 0.3312        | 0.0004 | 10   | 0.3662          |
+| 0.3696        | 0.0008 | 20   | 0.3296          |
+| 0.262         | 0.0012 | 30   | 0.3115          |
+| 0.3418        | 0.0016 | 40   | 0.3051          |
+| 0.3501        | 0.0020 | 50   | 0.3039          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:818b1f9e84cc241739191946c1953cb38bdcbbee4d6437973f2bfd497b3d8102
+oid sha256:4177ee615b19aad3a54ec880705dcbcd417d1f72b55865c2051605a9494da545
 size 50982842