Edit model card

hBERTv2_rte

This model is a fine-tuned version of gokuls/bert_12_layer_model_v2 on the GLUE RTE dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6896
  • Accuracy: 0.5487

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: 256
  • eval_batch_size: 256
  • seed: 10
  • distributed_type: multi-GPU
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 50
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.7231 1.0 10 0.7175 0.4549
0.702 2.0 20 0.7053 0.4729
0.6982 3.0 30 0.6976 0.4585
0.7008 4.0 40 0.7261 0.4657
0.7022 5.0 50 0.7142 0.4946
0.6867 6.0 60 0.6943 0.4801
0.6796 7.0 70 0.6896 0.5487
0.6614 8.0 80 0.7151 0.5162
0.6303 9.0 90 0.7244 0.5271
0.602 10.0 100 0.7570 0.4729
0.5761 11.0 110 0.7605 0.5379
0.5664 12.0 120 0.8160 0.5235

Framework versions

  • Transformers 4.26.1
  • Pytorch 1.14.0a0+410ce96
  • Datasets 2.10.1
  • Tokenizers 0.13.2
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.

Dataset used to train gokuls/hBERTv2_rte

Evaluation results