roberta-large-atomic-anion-1e-06-256

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

• Loss: 0.3194

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: 256
• seed: 42
• 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: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
0.4954	1.0	1152	0.4666
0.4156	2.0	2304	0.4036
0.3791	3.0	3456	0.3861
0.3586	4.0	4608	0.3600
0.3372	5.0	5760	0.3483
0.3232	6.0	6912	0.3380
0.3075	7.0	8064	0.3362
0.2964	8.0	9216	0.3308
0.2811	9.0	10368	0.3284
0.2733	10.0	11520	0.3240
0.2668	11.0	12672	0.3204
0.2608	12.0	13824	0.3210
0.2509	13.0	14976	0.3186
0.2426	14.0	16128	0.3173
0.2358	15.0	17280	0.3191
0.2307	16.0	18432	0.3205
0.232	17.0	19584	0.3149
0.2264	18.0	20736	0.3174
0.2199	19.0	21888	0.3199
0.2188	20.0	23040	0.3194

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.3.2
• Tokenizers 0.21.1