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/ef95acb3d3b15fe7_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 4bb9ce49-3066-4ab4-9235-6770b0be1c58
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 4bb9ce49-3066-4ab4-9235-6770b0be1c58
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 4bb9ce49-3066-4ab4-9235-6770b0be1c58
 
 This model is a fine-tuned version of [unsloth/Qwen2.5-Math-1.5B](https://huggingface.co/unsloth/Qwen2.5-Math-1.5B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: nan
 
 ## Model description
 
@@ -129,13 +131,18 @@ 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        | 0.0085 | 1    | nan             |
+| 0.0           | 0.0846 | 10   | nan             |
+| 0.0           | 0.1692 | 20   | nan             |
+| 0.0           | 0.2538 | 30   | nan             |
+| 0.0           | 0.3384 | 40   | nan             |
+| 0.0           | 0.4231 | 50   | nan             |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a33bd283f1f5db966a0fe5b86ca6b574fdfcddb1f78305f4f261060e925560a9
+oid sha256:44a2f15b0730dc8a2fccde323d96b4b68992f6ef37c8f6f6b664fb9cd5e8783f
 size 37070634