decompiler-distil-v2

This model is a fine-tuned version of Qwen/Qwen3-4B-Thinking-2507 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3561

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.4817	0.2255	100	0.3927
1.3607	0.4510	200	0.3683
1.2661	0.6764	300	0.3602
1.0269	0.9019	400	0.3561

Framework versions

• PEFT 0.18.0
• Transformers 4.57.3
• Pytorch 2.9.1+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1