english_to_taglish_3581

This model is a fine-tuned version of facebook/nllb-200-distilled-1.3B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9466

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: 1e-05
• train_batch_size: 64
• eval_batch_size: 16
• 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: cosine_with_restarts
• lr_scheduler_warmup_steps: 750
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
1.7717	1.0	45	1.5722
1.6939	2.0	90	1.5225
1.5634	3.0	135	1.3882
1.3812	4.0	180	1.2081
1.2163	5.0	225	1.1085
1.1352	6.0	270	1.0701
1.1098	7.0	315	1.0441
1.0361	8.0	360	1.0222
1.032	9.0	405	1.0028
0.9981	10.0	450	0.9887
0.9604	11.0	495	0.9773
0.9084	12.0	540	0.9677
0.8992	13.0	585	0.9594
0.8564	14.0	630	0.9508
0.8165	15.0	675	0.9477
0.777	16.0	720	0.9441
0.7341	17.0	765	0.9425
0.6906	18.0	810	0.9471
0.6756	19.0	855	0.9470
0.6751	20.0	900	0.9466

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1