KoMiniLLaDA-0.3B-QA

This model is a fine-tuned version of JuyeopDang/KoMiniLLaDA-0.3B-QA on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.0338

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• 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
• lr_scheduler_warmup_steps: 1000
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
3.0954	0.2298	5000	3.0173
3.059	0.4596	10000	3.0515
3.0471	0.6893	15000	2.9814
3.0143	0.9191	20000	2.9326
3.0074	1.1489	25000	3.0600
2.9786	1.3787	30000	2.8068
2.9969	1.6085	35000	2.8245
2.946	1.8382	40000	3.0338

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1