text_entailment_anli_large

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

• Loss: 0.3499
• Accuracy: 0.8698

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: 32
• eval_batch_size: 32
• 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1091	0.0565	200	1.0719	0.4349
0.878	0.1130	400	0.6520	0.7378
0.6166	0.1694	600	0.5196	0.8047
0.5353	0.2259	800	0.4664	0.8219
0.506	0.2824	1000	0.4422	0.8294
0.488	0.3389	1200	0.4499	0.8346
0.4308	0.3954	1400	0.4239	0.8445
0.4353	0.4518	1600	0.3890	0.8558
0.4102	0.5083	1800	0.4058	0.8461
0.4075	0.5648	2000	0.4076	0.8575
0.3974	0.6213	2200	0.3885	0.8588
0.4107	0.6778	2400	0.3499	0.8698
0.3868	0.7343	2600	0.3602	0.8673
0.3665	0.7907	2800	0.3607	0.8751
0.3768	0.8472	3000	0.3772	0.8670

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0