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: 8
 mlflow_experiment_name: /tmp/f8dd1b9a4074633c_train_data.json
 model_type: AutoModelForCausalLM
@@ -89,7 +89,7 @@ wandb_name: 72e3e03b-51b5-4716-9570-b72ad0323b1d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 72e3e03b-51b5-4716-9570-b72ad0323b1d
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -100,6 +100,8 @@ xformers_attention: null
 # 72e3e03b-51b5-4716-9570-b72ad0323b1d
 
 This model is a fine-tuned version of [MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4](https://huggingface.co/MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0000
 
 ## Model description
 
@@ -127,13 +129,17 @@ The following hyperparameters were used during training:
 - 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
+- training_steps: 34
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.0294 | 1    | 7.9631          |
+| No log        | 0.0294 | 1    | 7.9644          |
+| No log        | 0.2059 | 7    | 0.2206          |
+| 3.6996        | 0.4118 | 14   | 0.0001          |
+| 0.0005        | 0.6176 | 21   | 0.0000          |
+| 0.0005        | 0.8235 | 28   | 0.0000          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e17257a47de03d2922013b90ba37695e94ffdf9ea128f9b87a00cf39d1ee15a9
+oid sha256:7d9bc5ecc0655ec88a3e29abb8eb20bedc123d220d1fdff0d895b6021cf2069d
 size 608352466