yesj1234/jako_mbartLarge_100p_sup2

enko_mbartLarge_100p_sup2

This model is a fine-tuned version of facebook/mbart-large-50-many-to-many-mmt on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6417
• Bleu: 59.1835
• Gen Len: 15.7226

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: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2500
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.7676	1.0	43024	0.7125	55.2526	16.382
0.6349	2.0	86048	0.6547	58.202	15.9466
0.537	3.0	129072	0.6417	59.1835	15.7226
0.434	4.0	172096	0.6589	59.6194	15.702
0.3504	5.0	215120	0.7117	59.352	15.7454
0.2799	6.0	258144	0.7784	59.2034	15.6702

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1