results

This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.1152

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: 1
• eval_batch_size: 1
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
14.0328	1.0	586	13.4181
13.513	2.0	1172	10.5642
10.507	3.0	1758	5.6171
5.7516	4.0	2344	3.1528
2.8842	5.0	2930	2.4462
1.8802	6.0	3516	2.0028
1.5495	7.0	4102	1.7180
1.4116	8.0	4688	1.4359
1.3089	9.0	5274	1.3587
1.1416	10.0	5860	1.2989
1.1086	11.0	6446	1.2506
0.9502	12.0	7032	1.1806
0.9889	13.0	7618	1.1599
0.887	14.0	8204	1.1429
0.9452	15.0	8790	1.1311
0.8131	16.0	9376	1.1235
0.8805	17.0	9962	1.1191
0.8354	18.0	10548	1.1163
0.8135	19.0	11134	1.1153
0.8292	20.0	11720	1.1152

Framework versions

• PEFT 0.17.0
• Transformers 4.54.1
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.4