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/323546a4310179cb_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: b99d6895-46b0-40d9-83fd-c1c9c26d613d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: b99d6895-46b0-40d9-83fd-c1c9c26d613d
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 6d65b9b0-95e9-4290-b9d2-441d4803fa27
 
 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: 1.7592
 
 ## 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        | 0.0009 | 1    | 2.4728          |
+| 2.1751        | 0.0086 | 10   | 2.1014          |
+| 1.6291        | 0.0172 | 20   | 1.8043          |
+| 1.6754        | 0.0258 | 30   | 1.7723          |
+| 1.762         | 0.0344 | 40   | 1.7616          |
+| 1.8069        | 0.0430 | 50   | 1.7592          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82b6d404c57c2e94d87a3fed429a0782052227bce17ea8742d3f7477b658796b
+oid sha256:32fd15715a3c94407786c19653e199aeabb636e9d80e133db1ec47870685ab3d
 size 84047370