qwen2-sql-lora

This model is a fine-tuned version of Qwen/Qwen2-0.5B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6407

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: 0.0002
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use 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: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6799	0.3282	500	0.6805
0.6903	0.6564	1000	0.6626
0.6142	0.9846	1500	0.6526
0.6491	1.3124	2000	0.6463
0.6526	1.6406	2500	0.6428
0.6208	1.9688	3000	0.6407

Framework versions

• PEFT 0.17.0
• Transformers 4.55.2
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4