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: 8
 mlflow_experiment_name: /tmp/b50618247d909053_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 33ce9c8c-bca2-4c7a-816a-49b52dd1817c
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 33ce9c8c-bca2-4c7a-816a-49b52dd1817c
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 33ce9c8c-bca2-4c7a-816a-49b52dd1817c
 
 This model is a fine-tuned version of [Qwen/Qwen2-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2-1.5B-Instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 6.4040
 
 ## Model description
 
@@ -129,13 +131,17 @@ 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: 22
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0462 | 1    | 15.6474         |
+| No log        | 0.2312 | 5    | 14.1370         |
+| 13.9558       | 0.4624 | 10   | 10.0176         |
+| 13.9558       | 0.6936 | 15   | 7.1227          |
+| 7.7794        | 0.9249 | 20   | 6.4040          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c183baa9f947cded320be253d092bbb1b399df3a3ff62d9afddc67e066070784
+oid sha256:6ff074d634960c131a3132e5f2bfbc475ec8329ab2361b2a735f5311099e90a1
 size 37070634