bart-large-mnli_nli

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

• Loss: 3.1748
• Accuracy: 0.6146
• Precision Macro: 0.6164
• Recall Macro: 0.6143
• F1 Macro: 0.6144
• F1 Weighted: 0.6145

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Macro	Recall Macro	F1 Macro	F1 Weighted
0.9341	1.0	143	0.9033	0.5916	0.6003	0.5914	0.5927	0.5926
0.858	2.0	286	0.9499	0.5477	0.6120	0.5459	0.5366	0.5368
0.8298	3.0	429	0.9050	0.5849	0.5995	0.5868	0.5782	0.5775
0.7423	4.0	572	0.9457	0.6027	0.6128	0.6037	0.5964	0.5959
0.6525	5.0	715	0.9301	0.6009	0.6127	0.6002	0.6008	0.6008
0.5548	6.0	858	1.0490	0.6200	0.6258	0.6209	0.6164	0.6159
0.4609	7.0	1001	1.0379	0.6137	0.6159	0.6144	0.6131	0.6129
0.3427	8.0	1144	1.1565	0.6018	0.6050	0.6013	0.6017	0.6017
0.2528	9.0	1287	1.3466	0.5809	0.5906	0.5801	0.5788	0.5790
0.1635	10.0	1430	1.5895	0.6111	0.6163	0.6121	0.6090	0.6087
0.1308	11.0	1573	1.7222	0.6106	0.6118	0.6111	0.6104	0.6103
0.0883	12.0	1716	2.1606	0.6027	0.6052	0.6022	0.6019	0.6020
0.057	13.0	1859	2.2558	0.6093	0.6101	0.6093	0.6088	0.6087
0.0431	14.0	2002	2.4892	0.6169	0.6173	0.6168	0.6168	0.6168
0.0246	15.0	2145	2.8257	0.6155	0.6178	0.6153	0.6147	0.6147
0.0144	16.0	2288	2.8977	0.6169	0.6172	0.6168	0.6168	0.6168
0.0116	17.0	2431	3.0826	0.6173	0.6202	0.6170	0.6175	0.6175
0.0071	18.0	2574	3.1267	0.6164	0.6186	0.6160	0.6163	0.6163
0.0019	19.0	2717	3.1718	0.6142	0.6161	0.6138	0.6139	0.6140
0.0033	20.0	2860	3.1748	0.6146	0.6164	0.6143	0.6144	0.6145

Framework versions

• Transformers 4.55.0
• Pytorch 2.7.0+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4