metadata

license: mit
base_model: kennethge123/superglue_rte-gpt2
tags:
  - generated_from_trainer
datasets:
  - bigbench
metrics:
  - accuracy
model-index:
  - name: entailed_after_rte-gpt2
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: bigbench
          type: bigbench
          config: entailed_polarity
          split: validation
          args: entailed_polarity
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7142857142857143

entailed_after_rte-gpt2

This model is a fine-tuned version of kennethge123/superglue_rte-gpt2 on the bigbench dataset. It achieves the following results on the evaluation set:

Loss: 1.1865
Accuracy: 0.7143

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: 5e-05
train_batch_size: 4
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	30	0.9025	0.3571
No log	2.0	60	0.7155	0.5
No log	3.0	90	1.0014	0.2857
No log	4.0	120	0.9748	0.5714
No log	5.0	150	0.9511	0.5714
No log	6.0	180	1.0164	0.6429
No log	7.0	210	1.6015	0.5
No log	8.0	240	1.2833	0.6429
No log	9.0	270	1.0093	0.7857
No log	10.0	300	1.6339	0.6429
No log	11.0	330	1.3461	0.5714
No log	12.0	360	1.2949	0.6429
No log	13.0	390	1.6343	0.6429
No log	14.0	420	0.8418	0.8571
No log	15.0	450	0.6750	0.8571
No log	16.0	480	2.0221	0.6429
0.5929	17.0	510	0.7579	0.8571
0.5929	18.0	540	1.5713	0.7143
0.5929	19.0	570	1.0489	0.7143
0.5929	20.0	600	1.1865	0.7143

Framework versions

Transformers 4.37.0
Pytorch 1.13.1+cu117
Datasets 2.15.0
Tokenizers 0.15.2