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/8e03407abad422a8_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: f0ff9f09-e97a-426e-99a1-a97106892fe5
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: f0ff9f09-e97a-426e-99a1-a97106892fe5
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 69c9d7ac-1aa5-47f0-c44f-1bd557bec9da
 
 This model is a fine-tuned version of [tokyotech-llm/Llama-3-Swallow-8B-v0.1](https://huggingface.co/tokyotech-llm/Llama-3-Swallow-8B-v0.1) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.8388
 
 ## 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    | 2.8833          |
+| 2.8841        | 0.0019 | 10   | 2.6483          |
+| 2.1897        | 0.0038 | 20   | 2.0403          |
+| 1.9077        | 0.0057 | 30   | 1.9117          |
+| 1.962         | 0.0076 | 40   | 1.8478          |
+| 2.0557        | 0.0096 | 50   | 1.8388          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36006f4c31604f0483c52e9652f6613a7a113071994ef6522ec7feca2cb45224
+oid sha256:f351170a409f3e4689ec210139d352322bd53127d2ce5a09978b3cf854f0694c
 size 84047370