Wollaston/gelato-xlm-roberta-large

xlm-roberta-large

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0676
• Precision: 0.9471
• Recall: 0.9388
• F1: 0.9429
• Accuracy: 0.9884

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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_ratio: 0.2
• num_epochs: 18

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	10	1.8595	0.04	0.0035	0.0064	0.7470
No log	2.0	20	0.9639	0.0	0.0	0.0	0.7473
No log	3.0	30	0.5677	0.2834	0.2483	0.2647	0.8380
No log	4.0	40	0.2905	0.6278	0.7255	0.6732	0.9415
No log	5.0	50	0.1393	0.8413	0.8619	0.8515	0.9682
No log	6.0	60	0.0975	0.8906	0.8969	0.8937	0.9799
No log	7.0	70	0.0858	0.9032	0.8969	0.9	0.9822
No log	8.0	80	0.0708	0.9288	0.9126	0.9206	0.9841
No log	9.0	90	0.0944	0.9211	0.8986	0.9097	0.9817
No log	10.0	100	0.0636	0.9132	0.9196	0.9164	0.9863
No log	11.0	110	0.0544	0.9190	0.9318	0.9253	0.9879
No log	12.0	120	0.0585	0.9382	0.9283	0.9332	0.9878
No log	13.0	130	0.0578	0.9333	0.9301	0.9317	0.9878
No log	14.0	140	0.0600	0.9286	0.9318	0.9302	0.9876
No log	15.0	150	0.0656	0.9487	0.9371	0.9428	0.9883
No log	16.0	160	0.0675	0.9471	0.9388	0.9429	0.9884
No log	17.0	170	0.0669	0.9454	0.9388	0.9421	0.9886
No log	18.0	180	0.0676	0.9471	0.9388	0.9429	0.9884

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.1.1
• Tokenizers 0.22.1