metadata

tags:
  - generated_from_keras_callback
datasets:
  - snli
  - scitail
  - multi_nli
  - alisawuffles/WANLI
  - pietrolesci/nli_fever
  - facebook/anli
model-index:
  - name: electra-nli_finetuned
    results: []

electra-nli_finetuned

The model base is electra-small. It has been fine-tuned on: snli, scitail, wanli, mnli, fever_nli, anli.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

Training results

It achieved the following accuracy during training:
snli: 89.15
scitail: 90.08%
wanli: 67.84%
mnli: 81.95%
nli_fever: 74.14%
anli-r1_test: 46.60%
anli-r2_test: 42.50%
anli-r3_test: 43.08%

Framework versions

Transformers 4.31.0
TensorFlow 2.12.0
Tokenizers 0.13.3