small-vanilla-target-glue-rte

This model is a fine-tuned version of google/bert_uncased_L-4_H-512_A-8 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.8377
• Accuracy: 0.6390

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4103	6.41	500	1.1929	0.6318
0.0583	12.82	1000	2.3816	0.6029
0.0263	19.23	1500	2.7340	0.6065
0.0129	25.64	2000	2.8019	0.6354
0.015	32.05	2500	2.5423	0.6426
0.0168	38.46	3000	2.7661	0.6606
0.0112	44.87	3500	2.8761	0.6318
0.0119	51.28	4000	2.9552	0.6426
0.0163	57.69	4500	2.8377	0.6390

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu116
• Datasets 2.8.1.dev0
• Tokenizers 0.13.2