zoo-intent-results

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• F1-score: 0.9945
• Loss: 0.0260
• Accuracy: 0.9966
• F1 Macro: 0.9945
• F1 Weighted: 0.9966
• F1 Info: 0.9976
• F1 Routing: 0.9981
• F1 Smalltalk: 0.9861
• F1 Fallback: 0.9963

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: 2e-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
• num_epochs: 3

Training results

Training Loss	Epoch	Step	F1-score	Validation Loss	Accuracy	F1 Macro	F1 Weighted	F1 Info	F1 Routing	F1 Smalltalk	F1 Fallback
0.0827	1.0	8000	0.9879	0.0400	0.9927	0.9879	0.9928	0.9945	0.9959	0.9684	0.9928
0.0211	2.0	16000	0.9914	0.0330	0.9944	0.9914	0.9944	0.9959	0.9961	0.9793	0.9942
0.0142	3.0	24000	0.9945	0.0260	0.9966	0.9945	0.9966	0.9976	0.9981	0.9861	0.9963

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu126
• Datasets 4.1.1
• Tokenizers 0.22.1