THUDM/GLM-130B

Issues

• 训练log中的loss scale指的是什么？

  #221 opened 2 months ago by xliu99
  0

• Could you offer a download link with Chinese mainland mirror?

  #220 opened 8 months ago by GreekPanda
  0

• error about the GLM-130B’s model checkpoint

  #219 opened 9 months ago by sunpian1
  1

• 下载到一半就再也下不了了

  #218 opened 10 months ago by HaHaLiang666
  0

• 请各位大佬伸以援手，我想要在自己本地部署一个该模型，怎么在windows上进行部署？

  #217 opened a year ago by kangkangkangkkkk
  0

• 有用tensortRT-llm的docker环境跑通模型的吗？求助...

  #216 opened a year ago by dahaobenhao
  0

• 执行bash scripts/generate.sh --input-source interactive时出现的错误。大佬救救！

  #214 opened a year ago by Eternal-Yan
  1

• Clarification Request on GLM-130B Model Architecture and Licensing for Commercial Use

  #215 opened a year ago by JayLiangs
  0

• 8卡 fastertransformer 推理报错RuntimeError: [FT][ERROR] Assertion fail: /home/young.ruan/FasterTransformer/src/fastertransformer/th_op/glm/GlmOp.h:539

  #213 opened a year ago by rGitcy
  0

• RuntimeError: probability tensor contains either `inf`, `nan` or element < 0answers, answers_with_style, blanks = fill_blanks(raw_text, model, tokenizer, strategy)

  #212 opened a year ago by rGitcy
  0

• glm2-130B will it be made?

  #209 opened a year ago by yhyu13
  1

• 请问，课程链接在哪里？

  #210 opened a year ago by Stonesusu
  1

• Embedding Layer Gradient Shrink在哪里实现的？

  #191 opened 2 years ago by jiezhangGt
  1

• 模型效果很差，是什么原因呢？

  #186 opened 2 years ago by rchanggogogo
  6

• [ERROR] `bash scripts/generate.sh --input-source interactive` 报错

  #177 opened 2 years ago by SniperM99
  7

• FasterTransformer能否支持Glm6B呢

  #208 opened a year ago by sym19991125
  0

• 申请邮件收到的模型下载链接都失效了

  #207 opened a year ago by bixyz
  5

• 模型申请页面无法提交申请

  #205 opened 2 years ago by VSRacer
  1

• bash scripts/generate.sh --input-source interactive运行报错

  #184 opened 2 years ago by wudajun7509
  4

• 基于130B有chat版本开源的计划吗？

  #206 opened a year ago by ricosr
  0

• 如何使用FasterTransformer适配自己的模型

  #182 opened 2 years ago by ming-shy
  1

• 请问GLM可以在输出内容时，同时输出引用内容的来源吗？

  #204 opened 2 years ago by mike-2020
  0

• 内测的ChatGLM(https://chatglm.cn)使用感受还不如本地部署的chatGLM-6B量化模型，这是为啥?

  #192 opened 2 years ago by zhaochuninhefei
  1

• 6 cards inference

  #194 opened 2 years ago by wangheqi987
  1

• 模型并行集群怎么搭建

  #203 opened 2 years ago by ChenBinfighting1
  0

• GLM-130B文档中描述model weights，GPU内存需要260G，测试demo中实际测试总占用在240G左右，请问是什么原因

  #202 opened 2 years ago by zxs789
  0

• 每个token耗时呈脉冲式变化

  #201 opened 2 years ago by wangheqi987
  0

• 关于FT inference benchmark数据的疑问

  #200 opened 2 years ago by frankxyy
  0

• 训练目标

  #199 opened 2 years ago by shuangshuangguo
  0

• 关于docs/quantization.md中图片疑问

  #198 opened 2 years ago by M3Dade
  0

• 4*4090gpu for int4 model inference error

  #174 opened 2 years ago by sukibean163
  1

• [Question]GLM-130B模型有vocab文件吗？

  #195 opened 2 years ago by starkhu
  1

• GLM-130B 模型结构超参问题

  #196 opened 2 years ago by peiyingxin
  0

• FasterTransformer支持bf16推理吗

  #193 opened 2 years ago by benyang0506
  0

• GLM-130B如何使用lora微调

  #190 opened 2 years ago by ShaunHeNJU
  0

• 请问，GLM-130B有部署到DCU上的教程吗？

  #189 opened 2 years ago by guoxiaoyue111111
  0

• nvlink通信

  #188 opened 2 years ago by wangheqi987
  0

• 是不是chatglm与这个GLM-130b开源模型中间还有很多问题待解决？

  #178 opened 2 years ago by applepieiris
  2

• aria2的http_proxy和https_proxy报错

  #187 opened 2 years ago by Timaos123
  1

• 现在好像没有ChatGLM-130B开源吧？只有6B， 130B的不是Chat

  #183 opened 2 years ago by guotong1988
  1

• int4模型加载报错

  #185 opened 2 years ago by wudajun7509
  3

• RuntimeError: CUDA error: invalid device ordinal

  #181 opened 2 years ago by TranscenderNing
  1

• [HELP] 有人能分享一下量化好的int4 版本的模型吗?

  #179 opened 2 years ago by rchanggogogo
  0

• 想问一下作者，量化成int4 int8 之后为什么模型大小没有变化，都是240g

  #172 opened 2 years ago by GXKIM
  15

• 国内模型下载地址

  #176 opened 2 years ago by wangheqi987
  2

• 关于论文中bf16的一个疑问

  #180 opened 2 years ago by Saggressive
  0

• question: what does token mean here ?

  #175 opened 2 years ago by jiangying000
  0

• 请教

  #173 opened 2 years ago by ming-shy
  0

• https://tianqi.aminer.cn/ 天启官网合作咨询验证码打不开，请问如何联系商用

  #171 opened 2 years ago by sjtuzhaoxh
  1

• 为什么没有中文说明？

  #170 opened 2 years ago by fsy1215
  3