entailedpolarity_-bert-base-uncased-a2b1l0.001

This model was trained from scratch on the bigbench dataset. It achieves the following results on the evaluation set:

• Loss: 13.3548
• Accuracy: 0.875

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• 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	0.67	1	17.4094	0.2083
No log	2.0	3	15.8009	0.75
No log	2.67	4	15.3479	0.75
No log	4.0	6	14.4120	0.7917
No log	4.67	7	14.0308	0.7917
No log	6.0	9	13.0724	0.7917
No log	6.67	10	12.9712	0.7917
No log	8.0	12	12.8702	0.7917
No log	8.67	13	12.8670	0.7917
No log	10.0	15	13.0790	0.7917
No log	10.67	16	13.1837	0.8333
No log	12.0	18	13.2997	0.875
No log	12.67	19	13.3246	0.875
No log	13.33	20	13.3548	0.875

Framework versions

• Transformers 4.37.0
• Pytorch 1.13.1+cu117
• Datasets 2.15.0
• Tokenizers 0.15.2