End of training

README.md

@@ -66,7 +66,7 @@ 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/3e3baad8b42dc1fe_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: ae0c2f0c-99da-4e5b-885a-4f2b2b59bd8d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: ae0c2f0c-99da-4e5b-885a-4f2b2b59bd8d
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 7ecef951-5c3a-479a-8c5f-b38ac13c7d44
 
 This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B](https://huggingface.co/Qwen/Qwen2.5-1.5B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6381
 
 ## Model description
 
@@ -128,14 +130,18 @@ 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: 19
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0533 | 1    | 7.4367          |
+| No log        | 0.2133 | 4    | 7.3685          |
+| 7.5589        | 0.4267 | 8    | 5.4084          |
+| 6.183         | 0.64   | 12   | 1.2767          |
+| 2.2529        | 0.8533 | 16   | 0.6381          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb2c288fe8a393276d86a4bf1b81abcfd46f469faba1b2b3aa44dc6e80c28061
+oid sha256:c88b0c59de7ae936a71eb74b4394ffd79022bdc0058153b85d1f925a0005eca2
 size 37070634