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/9a7478f7dc0c77a4_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: 503b2cb9-e990-4d34-b81a-a3d784ed7e35
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 503b2cb9-e990-4d34-b81a-a3d784ed7e35
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # dae97f7e-c7bd-473b-b1ce-72e91441c5b2
 
 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: 0.5592
 
 ## 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        | 0.0000 | 1    | 1.7361          |
+| 1.3852        | 0.0002 | 10   | 1.4374          |
+| 0.8623        | 0.0004 | 20   | 0.7037          |
+| 0.6118        | 0.0005 | 30   | 0.5978          |
+| 0.5257        | 0.0007 | 40   | 0.5643          |
+| 0.5317        | 0.0009 | 50   | 0.5592          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:942085b1357479575a5409a99965c499b3ee253de6456c51553e81536470c69a
+oid sha256:5dd5e0a1c80064a9b1b67fce34ca2f0d6377a9cfaaf701679c8634c2019085e6
 size 80115210