End of training

README.md

@@ -64,7 +64,7 @@ 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/ddafe6aed2816972_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 0aab4601-8d83-4bf2-aca8-1e3e0578cf25
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 0aab4601-8d83-4bf2-aca8-1e3e0578cf25
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # cb04c128-780d-46f1-9306-db98b4c250ac
 
 This model is a fine-tuned version of [Korabbit/llama-2-ko-7b](https://huggingface.co/Korabbit/llama-2-ko-7b) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2307
 
 ## Model description
 
@@ -128,14 +130,19 @@ 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        | 0.0020 | 1    | 1.7533          |
+| 1.1502        | 0.0202 | 10   | 0.7086          |
+| 0.3534        | 0.0405 | 20   | 0.3280          |
+| 0.2478        | 0.0607 | 30   | 0.2478          |
+| 0.2272        | 0.0809 | 40   | 0.2336          |
+| 0.2033        | 0.1012 | 50   | 0.2307          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:88cf80b76f1ce9f74c2490d89cb601784f5e8be57b3db7257475a5a015319346
+oid sha256:9c9ed066a688e7bed91b12cf8c843a7e4412d3ab0e9f8efc9f75fd2d43b59725
 size 80115210