roberta-mc-3

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.5988
• Accuracy: 0.3

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-07
• 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
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6111	1.0	24	1.6061	0.4
1.6003	2.0	48	1.6062	0.4
1.6049	3.0	72	1.6062	0.4
1.5936	4.0	96	1.6059	0.4
1.6073	5.0	120	1.6057	0.4
1.6001	6.0	144	1.6055	0.3
1.5925	7.0	168	1.6052	0.3
1.5971	8.0	192	1.6050	0.3
1.597	9.0	216	1.6047	0.3
1.5956	10.0	240	1.6042	0.3
1.5882	11.0	264	1.6036	0.3
1.5944	12.0	288	1.6034	0.3
1.5941	13.0	312	1.6032	0.3
1.5941	14.0	336	1.6029	0.3
1.5825	15.0	360	1.6024	0.3
1.5817	16.0	384	1.6019	0.3
1.5922	17.0	408	1.6014	0.3
1.5915	18.0	432	1.6011	0.3
1.5822	19.0	456	1.6007	0.3
1.5967	20.0	480	1.6001	0.3
1.5887	21.0	504	1.5999	0.3
1.5905	22.0	528	1.5997	0.3
1.5828	23.0	552	1.5994	0.3
1.5851	24.0	576	1.5992	0.3
1.5789	25.0	600	1.5991	0.3
1.5797	26.0	624	1.5990	0.3
1.5845	27.0	648	1.5989	0.3
1.5992	28.0	672	1.5988	0.3
1.5791	29.0	696	1.5988	0.3
1.5785	30.0	720	1.5988	0.3

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.4
• Tokenizers 0.13.3