Skip to content

Issues: alibaba/Megatron-LLaMA

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

33 Open 22 Closed
33 Open 22 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
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

Llama 3 Support
#62 opened May 3, 2024 by john-theo
About batch_size
#61 opened Mar 16, 2024 by tszslovewanpu
sh LLaMA2_7B_standalone.sh
#60 opened Mar 14, 2024 by yangzhipeng1108
请问是否支持从0训练一个小规模的LLaMA模型，如：1B
#59 opened Mar 1, 2024 by liubo12
1
注意力权重转换问题
#58 opened Feb 19, 2024 by noob-ctrl
2
在模型转换权重时遇到了如下问题 Zarr-based strategies will not be registered because of missing packages
#57 opened Feb 5, 2024 by ZhangEnmao
使用distributed optimzer时grad_norm计算准确度的疑问
#56 opened Dec 27, 2023 by chivychao
1
LLaMAModel._causal_lm_process中的labels和logits对齐方法疑问
#53 opened Dec 8, 2023 by chivychao
3
Megatron-LM权重转hf格式
#52 opened Dec 2, 2023 by Yang-QW
4
问下readme中32机的吞吐对应的参数可以提供下吗，目前没有复现出来
#49 opened Nov 8, 2023 by jianzi123
5
求一份Serving的教程代码
#48 opened Nov 8, 2023 by xealml
1
hf权重转换代码小bug
#47 opened Nov 3, 2023 by yuanzhoulvpi2017
@thuhujin
INT4 量化的模型可以被Megatron-LLaMA支持吗？
#46 opened Oct 17, 2023 by Jeff123z
1
请问目前Megatron-LLaMA支持LLaMA2-70B的训练吗？
#45 opened Oct 17, 2023 by 13416157913
1
fp16的支持问题
#41 opened Oct 9, 2023 by XUWeijiang
1
OverlappedDistributedOptimizer 支持 pipeline parallelism > 1 和 data parallelism > 1 同时使用吗？
#37 opened Sep 26, 2023 by Baibaifan
8
对于不同参数模型，如何通过配置参数信息计算显存占用大小？
#36 opened Sep 26, 2023 by 13416157913
1
大家好，请教一个关于GLOBAL_BATCH_SIZE值计算的问题，希望大家不吝赐教。
#35 opened Sep 25, 2023 by 13416157913
1
训练LLaMA2-70B模型，4个节点，A800*80GB 设置8张量并行，1流水线并行，训练报错
#34 opened Sep 25, 2023 by 13416157913
多节点训练时使用nccl后端，在训练完后，保存检查点时报错
#32 opened Sep 22, 2023 by 13416157913
@thuhujin
1
Loss对齐
#31 opened Sep 22, 2023 by wuziyou199217
3
训练llama-30b模型报错是不支持llama-30b模型么？
#30 opened Sep 22, 2023 by 13416157913
请教一下，怎么感觉LLaMA2-7B模型单机A800*8*80G 用8张卡预训练TP4-PP1-DP2时间和TP1-PP1-DP8时间不合理
#24 opened Sep 18, 2023 by 13416157913
1
deepspeed+megatron+llama，请问作者有试过吗
#19 opened Sep 14, 2023 by Chandler-Bing
1
nccl通信边界问题？
#17 opened Sep 14, 2023 by Baibaifan
10
ProTip! Updated in the last three days: updated:>2024-05-12.