BART_HYDROGEN

This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1485

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: 8
• eval_batch_size: 16
• seed: 42
• 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
1.4785	1.1521	500	1.1364
1.2356	2.3041	1000	0.9821
1.111	3.4562	1500	0.8692
0.9972	4.6083	2000	0.7653
0.9087	5.7604	2500	0.6698
0.8306	6.9124	3000	0.5834
0.7472	8.0645	3500	0.4994
0.6657	9.2166	4000	0.4265
0.6187	10.3687	4500	0.3689
0.5562	11.5207	5000	0.3139
0.5121	12.6728	5500	0.2758
0.4661	13.8249	6000	0.2387
0.4268	14.9770	6500	0.2052
0.3932	16.1290	7000	0.1831
0.373	17.2811	7500	0.1661
0.3493	18.4332	8000	0.1558
0.3321	19.5853	8500	0.1494

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0