itserphan/grammar-correction-t5-lang8

results

This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6732
• Sacrebleu: 67.4689
• Rouge1: 0.8572
• Rouge2: 0.7151
• Rougel: 0.8458

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 128
• 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: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Sacrebleu	Rouge1	Rouge2	Rougel
0.7358	1.0	15250	0.6963	67.0067	0.8550	0.7105	0.8436
0.7441	2.0	30500	0.6810	67.2301	0.8560	0.7127	0.8446
0.7191	3.0	45750	0.6766	67.3703	0.8566	0.7139	0.8452
0.7033	4.0	61000	0.6755	67.4235	0.8570	0.7148	0.8457
0.7165	5.0	76250	0.6732	67.4689	0.8572	0.7151	0.8458

Framework versions

• Transformers 4.56.0
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0