segment_mask_token

This model is a fine-tuned version of zera09/custom-longt5-with-ts on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0534

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	72	4.5285
22.2051	2.0	144	0.8669
1.706	3.0	216	0.3556
1.706	4.0	288	0.1894
0.5941	5.0	360	0.1507
0.4187	6.0	432	0.1210
0.3811	7.0	504	0.0993
0.3811	8.0	576	0.0951
0.2151	9.0	648	0.0768
0.1378	10.0	720	0.0748
0.1378	11.0	792	0.0650
0.093	12.0	864	0.0660
0.0965	13.0	936	0.0632
0.0839	14.0	1008	0.0562
0.0839	15.0	1080	0.0576
0.0834	16.0	1152	0.0569
0.0795	17.0	1224	0.0560
0.0795	18.0	1296	0.0524
0.0752	19.0	1368	0.0539
0.1185	20.0	1440	0.0534

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0