ModelLink2/ascendspeed
gitee_code_template 6eb65d6dc9 baichuan13B模型
2023-09-28 10:30:20 +08:00
..
core up 2023-09-26 14:30:47 +08:00
data up 2023-09-26 14:30:47 +08:00
fp16_deprecated change megatron to ascendspeed 2023-06-10 21:26:01 +08:00
model baichuan13B模型 2023-09-28 10:30:20 +08:00
mpu up 2023-09-26 14:30:47 +08:00
ops Won't Import ascendspeed.ops if unneccssary 2023-08-16 09:35:10 +08:00
optimizer up 2023-09-26 14:30:47 +08:00
tokenizer up 2023-09-26 14:30:47 +08:00
__init__.py update ascendspeed/__init__.py. 2023-09-26 08:56:56 +00:00
arguments.py up 2023-09-26 14:30:47 +08:00
checkpointing.py up 2023-09-26 14:30:47 +08:00
data_classes.py up 2023-09-26 14:30:47 +08:00
enums.py change megatron to ascendspeed 2023-06-10 21:26:01 +08:00
error_utils.py up 2023-09-26 14:30:47 +08:00
global_vars.py up 2023-09-26 14:30:47 +08:00
indexer.py create megatron core 2023-07-24 15:00:57 +08:00
initialize.py up 2023-09-26 14:30:47 +08:00
learning_rates.py change megatron to ascendspeed 2023-06-10 21:26:01 +08:00
memory.py change megatron to ascendspeed 2023-06-10 21:26:01 +08:00
microbatches.py change megatron to ascendspeed 2023-06-10 21:26:01 +08:00
package_info.py change megatron to ascendspeed 2023-06-10 21:26:01 +08:00
text_generation_utils.py up 2023-09-26 14:30:47 +08:00
timers.py add ffts mode 2023-07-25 14:14:28 +08:00
training.py up 2023-09-26 14:30:47 +08:00
utils.py up 2023-09-26 14:30:47 +08:00