roberta-largemhr2004-atomic.anion.train.no1e-06-128

This model is a fine-tuned version of roberta-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3645

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: 512
• 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
0.6351	1.0	576	0.4929
0.4755	2.0	1152	0.4313
0.4406	3.0	1728	0.4136
0.4208	4.0	2304	0.4072
0.4089	5.0	2880	0.3982
0.3989	6.0	3456	0.3869
0.3913	7.0	4032	0.3837
0.3836	8.0	4608	0.3793
0.3766	9.0	5184	0.3793
0.3715	10.0	5760	0.3754
0.3669	11.0	6336	0.3726
0.3653	12.0	6912	0.3681
0.3594	13.0	7488	0.3697
0.3534	14.0	8064	0.3725
0.353	15.0	8640	0.3678
0.3509	16.0	9216	0.3651
0.3474	17.0	9792	0.3635
0.3467	18.0	10368	0.3676
0.3428	19.0	10944	0.3631
0.3408	20.0	11520	0.3668
0.3373	21.0	12096	0.3645
0.3397	22.0	12672	0.3645

Framework versions

• Transformers 4.51.2
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1