End of training

README.md

@@ -64,7 +64,7 @@ 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/cc257d59f46165df_train_data.json
 model_type: AutoModelForCausalLM
@@ -89,7 +89,7 @@ wandb_name: c84d9086-cc55-4202-87fa-830d96776ff3
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: c84d9086-cc55-4202-87fa-830d96776ff3
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -100,6 +100,8 @@ xformers_attention: null
 # 264cc8ce-80ad-156b-26e5-fcb377f6df87
 
 This model is a fine-tuned version of [katuni4ka/tiny-random-qwen1.5-moe](https://huggingface.co/katuni4ka/tiny-random-qwen1.5-moe) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.9278
 
 ## Model description
 
@@ -126,14 +128,19 @@ 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    | 11.9398         |
+| 11.9394       | 0.0090 | 10   | 11.9377         |
+| 11.9357       | 0.0180 | 20   | 11.9333         |
+| 11.9306       | 0.0270 | 30   | 11.9299         |
+| 11.9186       | 0.0360 | 40   | 11.9281         |
+| 11.9246       | 0.0450 | 50   | 11.9278         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f05678b8f9aa50753ddac3376aade2734bb5355add5535fb114490d53ef3168
+oid sha256:62a0061a5b06b5e78edf459a90da8758822472ae099bfc58261ca20885102bb5
 size 388266