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
..
.ipynb_checkpoints add nvidia megatron code 3 years ago
.pretrain_gpt2_distributed_2.6B.sh.swo add nvidia megatron code 3 years ago
debug_pretrain_gpt2_distributed_xxxM.sh add nvidia megatron code 3 years ago
evalPPL_gpt2_distributed.sh add nvidia megatron code 3 years ago
evaluate_zeroshot_gpt2.sh add nvidia megatron code 3 years ago
finetune_mnli_distributed.sh add nvidia megatron code 3 years ago
finetune_race_distributed.sh add nvidia megatron code 3 years ago
generate_text.sh add nvidia megatron code 3 years ago
generate_text_cmrc2018.sh add nvidia megatron code 3 years ago
merge_mp_bert.sh add nvidia megatron code 3 years ago
pretrain_bert.sh add nvidia megatron code 3 years ago
pretrain_bert_distributed.sh add nvidia megatron code 3 years ago
pretrain_gpt2.sh add nvidia megatron code 3 years ago
pretrain_gpt2_distributed.sh add nvidia megatron code 3 years ago
pretrain_gpt2_distributed_2.6B.sh add nvidia megatron code 3 years ago
pretrain_gpt2_distributed_345M.sh 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)