ERNIE-GRAM模型汇总#

下表汇总介绍了目前PaddleNLP支持的ERNIE-GRAM模型对应预训练权重。 关于模型的具体细节可以参考对应链接。

Pretrained Weight

Language

Details of the model

ernie-gram-zh

Chinese

12-layer, 768-hidden, 12-heads, 108M parameters. Trained on Chinese text.

ernie-gram-zh-finetuned-dureader-robust

Chinese

12-layer, 768-hidden, 12-heads, 108M parameters. Trained on Chinese text. Then finetuned on dreader-robust.