qwen-assamese-ner-lora

This model is a fine-tuned version of Qwen/Qwen3-0.6B on the assamese_ner_train dataset. It achieves the following results on the evaluation set:

• Loss: 0.0928

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: 0.0005
• train_batch_size: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 15.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.1607	1.0	1343	0.3104
0.1208	2.0	2686	0.1830
0.1098	3.0	4029	0.1470
0.1062	4.0	5372	0.1498
0.0557	5.0	6715	0.1195
0.0722	6.0	8058	0.1009
0.0418	7.0	9401	0.1032
0.0352	8.0	10744	0.0861
0.0255	9.0	12087	0.0861
0.015	10.0	13430	0.0812
0.0118	11.0	14773	0.0737
0.0123	12.0	16116	0.0865
0.0015	13.0	17459	0.0895
0.0012	14.0	18802	0.0876
0.0006	14.9894	20130	0.0928

Framework versions

• PEFT 0.14.0
• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1