Logo
Explore Help
Register Sign In
423A35C7/LLaMA-Factory
1
0
Fork 0
You've already forked LLaMA-Factory
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-08-05 05:02:50 +08:00
Code Issues Packages Projects Releases Wiki Activity
LLaMA-Factory/src/llamafactory
History
d da39715085 经过大量的增量预训练,进行对比试验,发现这个bug:llama3在预训练时使用的tokenizer.eos_toke是'<|end_of_text|>' ,这里在每条数据后面也得用这个,而不是'<|eot_id|>',否则很容易导致严重的性能下降
Former-commit-id: 6979f3f8480755604d8aea8164f6418126e094c5
2024-06-11 16:23:40 +08:00
..
api
update trainers
2024-06-06 18:45:49 +08:00
chat
rename files
2024-06-07 00:09:06 +08:00
data
经过大量的增量预训练,进行对比试验,发现这个bug:llama3在预训练时使用的tokenizer.eos_toke是'<|end_of_text|>' ,这里在每条数据后面也得用这个,而不是'<|eot_id|>',否则很容易导致严重的性能下降
2024-06-11 16:23:40 +08:00
eval
update evaluator
2024-06-10 23:56:00 +08:00
extras
set dev version
2024-06-11 00:50:53 +08:00
hparams
fix #4198
2024-06-11 15:38:38 +08:00
model
fix #4198
2024-06-11 15:38:38 +08:00
train
fix #4198
2024-06-11 15:38:38 +08:00
webui
add resume args in webui
2024-06-08 00:22:16 +08:00
__init__.py
better llamaboard
2024-05-29 23:55:38 +08:00
cli.py
fix #4137
2024-06-07 19:16:06 +08:00
launcher.py
support DDP in webui
2024-05-28 19:24:22 +08:00
Powered by Gitea Version: 23.8.0 Page: 40ms Template: 2ms
English
Bahasa Indonesia Deutsch English Español Français Gaeilge Italiano Latviešu Magyar nyelv Nederlands Polski Português de Portugal Português do Brasil Suomi Svenska Türkçe Čeština Ελληνικά Български Русский Українська فارسی മലയാളം 日本語 简体中文 繁體中文(台灣) 繁體中文(香港) 한국어
Licenses API