metadata

license: apache-2.0
base_model: google/electra-base-discriminator
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: electra_multiple_choice
    results: []

electra_multiple_choice

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

Loss: 1.4687
Accuracy: 0.55

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-06
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6127	1.0	12	1.6036	0.4
1.6001	2.0	24	1.5999	0.4
1.5916	3.0	36	1.5950	0.5
1.5849	4.0	48	1.5886	0.45
1.5793	5.0	60	1.5803	0.35
1.5642	6.0	72	1.5700	0.45
1.5416	7.0	84	1.5605	0.35
1.5106	8.0	96	1.5492	0.35
1.479	9.0	108	1.5357	0.35
1.443	10.0	120	1.5292	0.4
1.4114	11.0	132	1.5169	0.35
1.3505	12.0	144	1.5082	0.35
1.2985	13.0	156	1.5072	0.35
1.2493	14.0	168	1.4992	0.35
1.1686	15.0	180	1.4915	0.4
1.1307	16.0	192	1.4833	0.4
1.0781	17.0	204	1.4744	0.3
1.0179	18.0	216	1.4659	0.5
0.947	19.0	228	1.4576	0.5
0.9518	20.0	240	1.4522	0.5
0.8961	21.0	252	1.4535	0.45
0.8691	22.0	264	1.4598	0.45
0.8041	23.0	276	1.4512	0.45
0.7875	24.0	288	1.4568	0.45
0.7282	25.0	300	1.4601	0.45
0.7203	26.0	312	1.4595	0.5
0.6602	27.0	324	1.4606	0.5
0.6311	28.0	336	1.4687	0.55

Framework versions

Transformers 4.31.0
Pytorch 2.0.1+cu117
Datasets 2.14.4
Tokenizers 0.13.3