distilroberta-tcfd-disclosure

This model is a fine-tuned version of distilroberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.8681

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: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 5
• total_train_batch_size: 80
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	5	2.3837
2.3918	2.0	10	2.3787
2.3918	3.0	15	2.3704
2.3754	4.0	20	2.3623
2.3754	5.0	25	2.3396
2.2976	6.0	30	2.2599
2.2976	7.0	35	2.1095
2.0439	8.0	40	2.0184
2.0439	9.0	45	1.9059
1.6799	10.0	50	1.8469
1.6799	11.0	55	1.8089
1.2948	12.0	60	1.7263
1.2948	13.0	65	1.7250
0.9621	14.0	70	1.8106
0.9621	15.0	75	1.8073
0.7356	16.0	80	1.8681

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3