byt5-base-es_mim

This model is a fine-tuned version of google/byt5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5603
• Bleu: 8.1209
• Gen Len: 100.8969

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: 65
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	398	0.8136	0.014	19.0
1.246	2.0	796	0.7000	0.1109	19.0
0.8026	3.0	1194	0.6451	0.1504	19.0
0.7162	4.0	1592	0.6196	0.1224	19.0
0.7162	5.0	1990	0.5964	0.1486	19.0
0.6704	6.0	2388	0.5814	0.2053	19.0
0.6399	7.0	2786	0.5736	0.2048	19.0
0.6201	8.0	3184	0.5680	0.1875	19.0
0.6074	9.0	3582	0.5626	0.2047	19.0
0.6074	10.0	3980	0.5603	0.2074	19.0

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3