librarian-bot's picture
Librarian Bot: Add base_model information to model
c687fb5
|
raw
history blame
1.48 kB
metadata
license: cc-by-sa-4.0
tags:
  - generated_from_trainer
datasets:
  - cuad
base_model: nlpaueb/legal-bert-small-uncased
model-index:
  - name: legal-bert-small-uncased-cuad
    results: []

legal-bert-small-uncased-cuad

This model is a fine-tuned version of nlpaueb/legal-bert-small-uncased on the cuad dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0371

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: 2e-05
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
0.0565 1.0 11081 0.0411
0.0489 2.0 22162 0.0364
0.0409 3.0 33243 0.0371

Framework versions

  • Transformers 4.21.1
  • Pytorch 1.12.0+cu113
  • Datasets 2.4.0
  • Tokenizers 0.12.1