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/6e47aa8f2779b16d_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 162f6ddf-1d4b-4eba-b839-ad31c130e1d3
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 162f6ddf-1d4b-4eba-b839-ad31c130e1d3
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 5b25d07b-3d09-2b29-12b3-a737e3e32725
 
 This model is a fine-tuned version of [Qwen/Qwen1.5-0.5B-Chat](https://huggingface.co/Qwen/Qwen1.5-0.5B-Chat) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.4831
 
 ## Model description
 
@@ -128,14 +130,19 @@ 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        | 0.0000 | 1    | 2.7575          |
+| 2.7408        | 0.0003 | 10   | 2.6381          |
+| 2.6363        | 0.0007 | 20   | 2.5265          |
+| 2.5836        | 0.0010 | 30   | 2.4935          |
+| 2.588         | 0.0013 | 40   | 2.4848          |
+| 2.5283        | 0.0016 | 50   | 2.4831          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a960a2839f5fc42b3a457f3e7fb1af05452b5a821105f87bf2b1de666884628b
+oid sha256:22871d61a3672dd01415cb4aa685519d835dd63ad6cbe944f84236fdae6d7756
 size 15259594