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/17a9ad7cd40f3339_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: 3914fcfc-706c-4a02-ab6b-4cd0cbe61b21
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 3914fcfc-706c-4a02-ab6b-4cd0cbe61b21
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 3914fcfc-706c-4a02-ab6b-4cd0cbe61b21
 
 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: 1.4672
 
 ## 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        | 0.0006 | 1    | 1.7580          |
+| 1.7088        | 0.0057 | 10   | 1.7292          |
+| 1.3906        | 0.0114 | 20   | 1.5854          |
+| 1.629         | 0.0171 | 30   | 1.5051          |
+| 1.4302        | 0.0227 | 40   | 1.4721          |
+| 1.6478        | 0.0284 | 50   | 1.4672          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:063f2c95aafaf09fb78c82bc16b19a691b2ddbf16f4cba005eab2d07d6b73bda
+oid sha256:7992ea08752e72b223753ec67a158f85a4570047e6825d6399e4d259675fca8e
 size 80115210