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: 8
 mlflow_experiment_name: /tmp/64489f5a946bc196_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 4e2f986c-e920-4569-9f41-c0f572d9a01f
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 4e2f986c-e920-4569-9f41-c0f572d9a01f
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 4e2f986c-e920-4569-9f41-c0f572d9a01f
 
 This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3464
 
 ## Model description
 
@@ -128,13 +130,18 @@ 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: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0159 | 1    | 14.7175         |
+| 11.7539       | 0.1590 | 10   | 7.0242          |
+| 3.5086        | 0.3181 | 20   | 0.4485          |
+| 0.5005        | 0.4771 | 30   | 0.3772          |
+| 0.3881        | 0.6362 | 40   | 0.3639          |
+| 0.3583        | 0.7952 | 50   | 0.3464          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65dc128bfcceaa61cc98076f1e9175126725923b57d0314b913e24ceed104182
+oid sha256:fd0ba936c48b9d37c7354faadb3e4c27e45ba01e9707686851befd0a53c1749b
 size 80881450