nllb-200-distilled-1.3B-finetuned-finetuned-finetuned-finetuned

This model is a fine-tuned version of KevinKibe/nllb-200-distilled-1.3B-finetuned-finetuned-finetuned on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3966
• Rouge: 0.2713
• Gen Len: 22.5

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge	Gen Len
0.0032	200.0	200	0.4363	0.2835	22.0
0.0003	400.0	400	0.3573	0.2713	22.5
0.0	600.0	600	0.3756	0.2762	19.5
0.0	800.0	800	0.3805	0.2762	19.5
0.0	1000.0	1000	0.3809	0.2762	19.5
0.0	1200.0	1200	0.3848	0.2762	19.5
0.0	1400.0	1400	0.3964	0.2713	22.5
0.0	1600.0	1600	0.3961	0.2713	22.5
0.0	1800.0	1800	0.3964	0.2713	22.5
0.0	2000.0	2000	0.3966	0.2713	22.5

Framework versions

• Transformers 4.39.2
• Pytorch 2.2.2+cu121
• Datasets 2.21.0
• Tokenizers 0.15.2