bert-base-uncased-finetuned-cosmos

This model is a fine-tuned version of bert-base-uncased on the cosmos_qa dataset. It achieves the following results on the evaluation set:

• Loss: 3.1272
• Accuracy: 0.5752

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: 4
• 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	Accuracy
1.0246	1.0	6316	1.0667	0.5715
0.632	2.0	12632	1.5163	0.5926
0.2341	3.0	18948	3.1272	0.5752

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2