byt5-small-wikipron-eng-latn-in-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: 0.1796
• Per: 0.3243
• Gen Len: 16.3969

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
2.2292	1.0	233	0.3751	0.4995	16.0446
0.387	2.0	466	0.2547	0.4174	16.3408
0.2887	3.0	699	0.2157	0.362	16.3243
0.2459	4.0	932	0.2019	0.3432	16.3488
0.2224	5.0	1165	0.1925	0.3339	16.3654
0.207	6.0	1398	0.1868	0.328	16.3744
0.1956	7.0	1631	0.1816	0.3276	16.3906
0.1872	8.0	1864	0.1815	0.3241	16.3957
0.1823	9.0	2097	0.1812	0.3244	16.3927
0.1778	10.0	2330	0.1796	0.3243	16.3969

Framework versions

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