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/4a8c1f73129d119d_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: 7f565ec4-2541-4096-ace4-5e60ccdc0e69
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 7f565ec4-2541-4096-ace4-5e60ccdc0e69
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 2eae1142-3da5-eb6a-b41a-89347fe2ed84
 
 This model is a fine-tuned version of [elyza/Llama-3-ELYZA-JP-8B](https://huggingface.co/elyza/Llama-3-ELYZA-JP-8B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6013
 
 ## 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        | 0.0002 | 1    | 0.8827          |
+| 0.7633        | 0.0025 | 10   | 0.7921          |
+| 0.5986        | 0.0049 | 20   | 0.6489          |
+| 0.5604        | 0.0074 | 30   | 0.6137          |
+| 0.6725        | 0.0099 | 40   | 0.6030          |
+| 0.6081        | 0.0123 | 50   | 0.6013          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6edfbc336222f2007904183eb5fa80879b43769803d0713401cab847d101902c
+oid sha256:d734dc12d44aee3ce3f85ba3ad2ee7a0186bb211cd0d7f4b139061d7edeb9aec
 size 84047370