electra-large

This model is a fine-tuned version of google/electra-large-discriminator on the cuad dataset for the question-answering task.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

Contract Understanding Atticus Dataset (CUAD) is an extractive question-answering dataset on legal contracts proposed by the Atticus Project, a non-profit organization of legal experts, designed with the help of many experts in the legal field.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 1e-05
  • train_batch_size: 8
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 6.0

Training results

Framework versions

  • Transformers 4.27.4
  • Pytorch 2.0.0+cu117
  • Datasets 2.11.0
  • Tokenizers 0.13.2
Downloads last month
8
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train frankkuete/electra-large-cuad-qa