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
..
openwebtext add nvidia megatron code 3 years ago
create_doc_index.py add nvidia megatron code 3 years ago
file_iter.py add nvidia megatron code 3 years ago
generate_samples_gpt2.py add nvidia megatron code 3 years ago
gpu_mem_track.py add nvidia megatron code 3 years ago
linter.py add nvidia megatron code 3 years ago
merge_mp_partitions.py add nvidia megatron code 3 years ago
pre_process_chinese.py add nvidia megatron code 3 years ago
preprocess_data.py add nvidia megatron code 3 years ago
preprocess_data_ML.py add nvidia megatron code 3 years ago
tokenization_jieba.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)