You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 
zhanghangit bf716fb7c0 add nvidia megatron code 3 years ago
..
__init__.py add nvidia megatron code 3 years ago
bert_model.py add nvidia megatron code 3 years ago
classification.py add nvidia megatron code 3 years ago
distributed.py add nvidia megatron code 3 years ago
fused_bias_gelu.py add nvidia megatron code 3 years ago
fused_softmax.py add nvidia megatron code 3 years ago
gpt2_model.py add nvidia megatron code 3 years ago
language_model.py add nvidia megatron code 3 years ago
multiple_choice.py add nvidia megatron code 3 years ago
realm_model.py add nvidia megatron code 3 years ago
transformer.py add nvidia megatron code 3 years ago
utils.py add nvidia megatron code 3 years ago

使用100G中文高质量语料,128张V100,训练的中文Megatron模型,参数量2.6B

Python C++ Shell Cuda TeX other

Contributors (2)