byt5-small-wikipron-eng-latn-nz-broad

This model is a fine-tuned version of google/byt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1880
• Per: 0.3259
• Gen Len: 16.3063

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: 0.0002
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Per	Gen Len
2.2173	1.0	235	0.3860	0.5095	15.7391
0.3895	2.0	470	0.2617	0.4065	16.1533
0.2907	3.0	705	0.2238	0.3594	16.2184
0.2479	4.0	941	0.2088	0.3451	16.2767
0.2274	5.0	1176	0.2000	0.3362	16.2672
0.2114	6.0	1411	0.1949	0.3366	16.3016
0.1997	7.0	1646	0.1935	0.3333	16.3086
0.1909	8.0	1882	0.1899	0.327	16.307
0.1854	9.0	2117	0.1882	0.3261	16.3109
0.1811	9.99	2350	0.1880	0.3259	16.3063

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu117
• Datasets 2.12.0
• Tokenizers 0.13.3