Skip to content

Issues: yangjianxin1/Firefly

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

198 Open 80 Closed
198 Open 80 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

qwen2 pretrain loss非常大
#301 opened Sep 28, 2024 by ucaslei
目前调用unsloth是只能支持单卡吗
#300 opened Sep 19, 2024 by MinusOne2
关于参数remove_unused_columns = false
#299 opened Sep 8, 2024 by Kenneth0901
template中的system_format字段是必须的吗？对于微调来说，这个字段是不是可以不用，从节省输入token数量的角度来说
#298 opened Sep 2, 2024 by Chris-Mraz
设置了双卡，却信息任然显示单卡。怎么回事？
#297 opened Aug 22, 2024 by youzihaha
多卡load模型OOM
#296 opened Aug 19, 2024 by TonyUSTC
知识蒸馏
#295 opened Aug 12, 2024 by zjjznw123
loss一直为0
#293 opened Aug 2, 2024 by chk4991
1
用qwen1.5 infer的时候他就疯狂写代码
#292 opened Jul 31, 2024 by ExeCuteRunrunrun
有支持华为昇腾npu的计划吗
#291 opened Jul 30, 2024 by bestRiven
如何在训练语料中添加system
#290 opened Jul 26, 2024 by louyuanyuan053709
1
llama 3.1, DCLM支持
#288 opened Jul 24, 2024 by WinterStraw
请问有支持GLM4-9B-Chat SFT 的计划吗？
#287 opened Jul 23, 2024 by Candy555
category 这个字段在推理的时候怎么用
#286 opened Jul 17, 2024 by louyuanyuan053709
有支持internvl 的计划么？
#284 opened Jul 11, 2024 by moyans
请问我运行“python train.py --train_args_file train_args/sft/qlora/baichuan-7b-sft-qlora.json”指令，为什么报一下错误呀？
#283 opened Jul 9, 2024 by lh5533223
1
关于蒸馏训练的代码
#282 opened Jul 8, 2024 by rattlesnakey
4
训练baichuan-13B qlora, 发现loss查看为空，怎么解决
#281 opened Jul 6, 2024 by lovegit2021
目前unsloth是不是不支持qwen2-moe结构？
#280 opened Jul 4, 2024 by xiaoer2498
微调Qwen2-1.5B-Instruct，loss始终是0
#279 opened Jul 3, 2024 by frederichen01
7
DPO的训练能否像SFT一样mask掉user，只计算assistant的loss
#278 opened Jul 2, 2024 by charliedream1
5
现在是否支持微调后大模型的单机多卡推理。
#277 opened Jul 2, 2024 by SunRise-Star
有计划支持deepseek-coder-v2-lite的微调吗
#276 opened Jun 30, 2024 by fengyang95
可以支持 llava 微调吗？
#275 opened Jun 29, 2024 by vincent507cpu
expected string or bytes-like object
#274 opened Jun 28, 2024 by sankexin
ProTip! Add no:assignee to see everything that’s not assigned.