roberta-large-nsp-1000000-1e-06-32-negcommonsensebalanced-1e-06-64

This model is a fine-tuned version of mhr2004/roberta-large-nsp-1000000-1e-06-32 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3364

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-06
• train_batch_size: 256
• eval_batch_size: 1024
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
0.5348	1.0	795	0.4701
0.4543	2.0	1590	0.4139
0.4173	3.0	2385	0.3899
0.3926	4.0	3180	0.3784
0.3702	5.0	3975	0.3663
0.3578	6.0	4770	0.3554
0.3472	7.0	5565	0.3466
0.3334	8.0	6360	0.3467
0.3206	9.0	7155	0.3425
0.3114	10.0	7950	0.3408
0.3033	11.0	8745	0.3340
0.2948	12.0	9540	0.3339
0.2858	13.0	10335	0.3385
0.2829	14.0	11130	0.3333
0.2775	15.0	11925	0.3334
0.2652	16.0	12720	0.3367
0.2645	17.0	13515	0.3364

Framework versions

• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0