570e512
1
2
3
4
5
6
learning rate: 2e-5 training epochs: 3 batch size: 64 seed: 87 model: bert-base-uncased trained on MNLI which is converted into two-way nli classification (predict entailment or not-entailment class)
