metadata

language:
  - en
license: apache-2.0
datasets:
  - glue
metrics:
  - accuracy
model-index:
  - name: gpt2-finetuned-cola
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE COLA
          type: glue
          args: cola
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.77756

gpt2-finetuned-cola

This model is gpt2 fine-tuned on GLUE CoLA dataset. It acheives the following results on the validation set. Though MCC is used for CoLA dataset, accuracy is reported here

Accuracy: 0.77756

Model Details

GPT-2 is a transformers model pretrained on a very large corpus of English data in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was trained to guess the next word in sentences. However, it acheives very good results on Text Classification tasks.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 3e-4
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: epsilon=1e-08
num_epochs: 5.0

Training results

Epoch	Training Loss	Validation Loss	Training Accuracy	Validation Accuracy
1	0.65824	0.55883	0.67957	0.70086
2	0.51774	0.53402	0.74927	0.75264
3	0.44502	0.50641	0.79172	0.77756
4	0.38996	0.54241	0.82645	0.77469
5	0.35218	0.55088	0.84516	0.77277