t5-base-turkish-reincames-hukuk-qlora-v23

This model is a fine-tuned version of Turkish-NLP/t5-efficient-base-turkish on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0842

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: 0.0003
• train_batch_size: 16
• 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: 15

Training results

Training Loss	Epoch	Step	Validation Loss
2.8961	1.0	835	1.5937
1.8123	2.0	1670	1.4405
1.6381	3.0	2505	1.3488
1.5931	4.0	3340	1.2910
1.4777	5.0	4175	1.2431
1.417	6.0	5010	1.2103
1.4019	7.0	5845	1.1724
1.3517	8.0	6680	1.1541
1.3031	9.0	7515	1.1386
1.2741	10.0	8350	1.1180
1.2515	11.0	9185	1.1040
1.2302	12.0	10020	1.0992
1.1981	13.0	10855	1.0897
1.2079	14.0	11690	1.0874
1.2021	15.0	12525	1.0842

Framework versions

• PEFT 0.14.0
• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0