en-vi-envit5_TMG301

This model is a fine-tuned version of VietAI/envit5-translation on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1307
• Bleu: 17.4518

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
No log	0.3	5000	1.0773	17.4258
No log	0.6	10000	1.0833	17.3553
No log	0.9	15000	1.0725	17.4577
No log	1.2	20000	1.1025	17.2140
No log	1.5	25000	1.0993	17.4008
No log	1.8	30000	1.0939	17.6789
No log	2.1	35000	1.1323	17.2096
No log	2.4	40000	1.1322	17.2425
No log	2.7	45000	1.1307	17.4518

Framework versions

• Transformers 4.38.2
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.15.2