nli_model_results

This model is a fine-tuned version of typeform/distilbert-base-uncased-mnli on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.9139
• Accuracy: 0.805

Model description

The Model is a better version of uncased-mnli for the evaluation of

1. Entailment
2. Contradiction
3. Neutral Sentences based on theory.

Intended uses & limitations

Can we used with Hallucination detector Pipelines

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: 64
• eval_batch_size: 64
• 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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	157	1.6075	0.808
No log	2.0	314	1.7477	0.811
No log	3.0	471	1.9139	0.805

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.3
• Tokenizers 0.22.2