IndoT5-LoRA-R32-LR0.0003-Ep20

This model is a fine-tuned version of Wikidepia/IndoT5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4510

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
0.866	1.0	188	0.8142
0.5507	2.0	376	0.5509
0.5726	3.0	564	0.5266
0.5125	4.0	752	0.5234
0.5742	5.0	940	0.5098
0.5129	6.0	1128	0.5091
0.4981	7.0	1316	0.4601
0.4559	8.0	1504	0.4538
0.427	9.0	1692	0.4460
0.414	10.0	1880	0.4432
0.4007	11.0	2068	0.4467
0.3969	12.0	2256	0.4369
0.4267	13.0	2444	0.4397
0.3928	14.0	2632	0.4405
0.3992	15.0	2820	0.4510

Framework versions

• PEFT 0.18.1
• Transformers 4.57.6
• Pytorch 2.10.0+cu128
• Datasets 4.5.0
• Tokenizers 0.22.2