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/e7bc7844942102c2_train_data.json
 model_type: AutoModelForCausalLM
@@ -89,7 +89,7 @@ wandb_name: 44997448-2c94-4717-9608-bca017a9f721
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 44997448-2c94-4717-9608-bca017a9f721
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -100,6 +100,8 @@ xformers_attention: null
 # 7250e8f6-e6de-40f0-abf5-2d0a8225e9af
 
 This model is a fine-tuned version of [oopsung/llama2-7b-koNqa-test-v1](https://huggingface.co/oopsung/llama2-7b-koNqa-test-v1) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.7697
 
 ## Model description
 
@@ -126,14 +128,19 @@ 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        | 0.0002 | 1    | 2.9332          |
+| 2.7993        | 0.0018 | 10   | 2.9069          |
+| 2.7949        | 0.0036 | 20   | 2.8183          |
+| 2.7626        | 0.0054 | 30   | 2.7879          |
+| 2.6892        | 0.0072 | 40   | 2.7727          |
+| 2.7579        | 0.0089 | 50   | 2.7697          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c580747952a55d9585f8c2474fbd150da866a0c621e043d9f472a3a5ccb9273
+oid sha256:d9d6010f3b67b866520f2b8e6a11787482a9a0561a24166a7624acd70ba6f807
 size 80115210