refinement-finetuned-mnli-2

This model is a fine-tuned version of mfreihaut/refinement-finetuned-mnli-1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0242

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	303	0.3730
1.1146	2.0	606	0.9860
1.1146	3.0	909	0.7304
1.0018	4.0	1212	0.6386
1.0045	5.0	1515	0.4228
1.0045	6.0	1818	0.6769
0.9618	7.0	2121	0.3008
0.9618	8.0	2424	0.4496
0.964	9.0	2727	0.1826
0.9586	10.0	3030	0.0367
0.9586	11.0	3333	0.1811
1.0467	12.0	3636	0.1352
1.0467	13.0	3939	0.0612
1.0047	14.0	4242	0.1702
1.0012	15.0	4545	0.0622
1.0012	16.0	4848	0.7077
1.0514	17.0	5151	0.2146
1.0514	18.0	5454	0.5531
0.9389	19.0	5757	1.2304
0.9229	20.0	6060	0.6252
0.9229	21.0	6363	0.6844
0.9334	22.0	6666	0.5663
0.9334	23.0	6969	0.9912
0.9312	24.0	7272	0.3112
0.8971	25.0	7575	0.4511
0.8971	26.0	7878	0.3860
0.9022	27.0	8181	0.5904
0.9022	28.0	8484	0.4710
0.7568	29.0	8787	0.8233
0.6753	30.0	9090	0.6951
0.6753	31.0	9393	0.6363
0.5802	32.0	9696	0.8018
0.5802	33.0	9999	0.9381
0.5323	34.0	10302	0.9941
0.5218	35.0	10605	0.9418
0.5218	36.0	10908	0.9236
0.4558	37.0	11211	0.4542
0.4247	38.0	11514	0.9279
0.4247	39.0	11817	0.9567
0.43	40.0	12120	0.8077
0.43	41.0	12423	0.9595
0.352	42.0	12726	0.9189
0.3393	43.0	13029	0.8762
0.3393	44.0	13332	1.0505
0.316	45.0	13635	0.9273
0.316	46.0	13938	1.0716
0.2983	47.0	14241	1.0084
0.2503	48.0	14544	1.1027
0.2503	49.0	14847	1.0478
0.2462	50.0	15150	1.0242

Framework versions

• Transformers 4.22.2
• Pytorch 1.10.0
• Datasets 2.5.1
• Tokenizers 0.12.1