prn222-mt5-base

This model is a fine-tuned version of PhamNghia123/prn222-mt5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.5223

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• 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: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss
2.9490	1.0	25	2.7538
2.8012	2.0	50	2.8785
3.1793	3.0	75	2.7940
2.8507	4.0	100	2.8119
2.9249	5.0	125	2.7168
2.9049	6.0	150	2.7252
3.6162	7.0	175	2.6183
4.2384	8.0	200	2.4568
3.7363	9.0	225	2.4137
3.6711	10.0	250	2.4603
3.4256	11.0	275	2.4526
3.6470	12.0	300	2.4630
3.3000	13.0	325	2.4976
3.0273	14.0	350	2.5205
3.4287	15.0	375	2.5223

Framework versions

• Transformers 5.9.0
• Pytorch 2.11.0+cu128
• Datasets 4.8.5
• Tokenizers 0.22.2