opt-small-sw

This model is a fine-tuned version of facebook/opt-125m on the SwitchBoard dataset. It achieves the following results on the evaluation set:

  • Loss: 2.7166
  • Accuracy: 0.4663

Model description

More details soon.

Intended uses & limitations

More details soon.

Training and evaluation data

More details soon.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 7
  • eval_batch_size: 7
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20.0

Training results

Framework versions

  • Transformers 4.26.0
  • Pytorch 1.13.1
  • Datasets 2.9.0
  • Tokenizers 0.13.0.dev0
Downloads last month
7
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.