roberta-base_environmental_claims

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

• Loss: 0.4015

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
0.3333	1.0	133	0.4844
0.3429	2.0	266	0.4015
0.3122	3.0	399	0.6850
0.1914	4.0	532	0.6587
0.2417	5.0	665	0.5776
0.1458	6.0	798	0.6259
0.1755	7.0	931	0.6392
0.0664	8.0	1064	0.8447
0.0299	9.0	1197	0.7466
0.138	10.0	1330	0.6908
0.0935	11.0	1463	0.8548
0.0002	12.0	1596	0.7000
0.119	13.0	1729	1.2066
0.0001	14.0	1862	1.1844
0.0003	15.0	1995	0.7856
0.0001	16.0	2128	0.8803
0.0	17.0	2261	0.9005
0.0	18.0	2394	0.9338
0.0	19.0	2527	0.9946
0.1325	20.0	2660	0.7678
0.0001	21.0	2793	0.5737
0.0005	22.0	2926	1.0396
0.0855	23.0	3059	0.7681
0.0	24.0	3192	1.0212
0.0856	25.0	3325	1.0275
0.0	26.0	3458	0.7531
0.0	27.0	3591	0.7814
0.0	28.0	3724	0.8142
0.0	29.0	3857	1.4194
0.0	30.0	3990	1.1039
0.0	31.0	4123	1.2294
0.0	32.0	4256	1.1357
0.0	33.0	4389	0.9818
0.0	34.0	4522	0.9881
0.0	35.0	4655	0.9947
0.0	36.0	4788	1.0028
0.0	37.0	4921	1.0100
0.0	38.0	5054	1.0038
0.0	39.0	5187	1.0115
0.1009	40.0	5320	1.0044
0.0	41.0	5453	1.5468
0.1724	42.0	5586	1.0918
0.0	43.0	5719	1.4838
0.0	44.0	5852	1.3071
0.0	45.0	5985	1.3350
0.0	46.0	6118	1.3400
0.0	47.0	6251	1.3331
0.0	48.0	6384	1.3265
0.0	49.0	6517	1.2641
0.0	50.0	6650	1.2638

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.0.2
• Tokenizers 0.19.1