Search results

Thank you very much for your code. Do you have the code for how to count the memory in Table 1? Thank you very much

Status: Open.

#80 In OpenLMLab/LOMO;

· LiZhangMing opened on Nov 28, 2024

Serious conclusion: LOMO does not significantly reduce GPU memory usage！

Status: Open.

#72 In OpenLMLab/LOMO;

· misonsky opened on Jan 3, 2024

实测LOMO++ deepspeed zero2 7b qlora llama 显存占用 感觉比正常的qlora + deepspeed zero2 显存占用大1倍

Status: Open.

#70 In OpenLMLab/LOMO;

· zlh1992 opened on Dec 26, 2023

Can you provide detailed dependency versions?

Status: Open.

#69 In OpenLMLab/LOMO;

· misonsky opened on Dec 23, 2023

adalomo在使用chatglm2模型出现错误

Status: Open.

#68 In OpenLMLab/LOMO;

· JorunoJobana opened on Nov 7, 2023

adalomo 试图在 llama 2 70b 模型训练时出现 NCCL communicator 这类超时的错误

Status: Open.

#67 In OpenLMLab/LOMO;

· HappyLynn opened on Nov 7, 2023

eval environment for opencompass

Status: Open.

#65 In OpenLMLab/LOMO;

· KaiLv69 opened on Oct 31, 2023

Runtime error on 2nd epoch, trying lora only

Status: Open.

#56 In OpenLMLab/LOMO;

· wasifferoze opened on Oct 3, 2023

a bug found in save_model of LOMOTrainer

Status: Open.

#54 In OpenLMLab/LOMO;

· DingQiang2018 opened on Aug 30, 2023

LlaMA-7B + LoRA在16GB的V100上OOM

Status: Open.

#53 In OpenLMLab/LOMO;

· zhenqincn opened on Aug 30, 2023

Customized loss value

Status: Open.

#52 In OpenLMLab/LOMO;

· ZN1010 opened on Aug 22, 2023

请问自定义Dataset只能是classification的数据集吗

Status: Open.

#51 In OpenLMLab/LOMO;

· Tonystark64 opened on Jul 31, 2023