Commits

Breadcrumbs

History for

on425efee

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 8, 2023

• Merge pull request ztxz16#362 from wangyumu/add-stop-tokens-ids

  Show description for 737a58a
  ztxz16

  authored
  737a58a

• Merge pull request ztxz16#359 from purpleroc/master

  Show description for f97fff8
  ztxz16

  authored
  f97fff8

• 支持ChatGLM3模型的Prompt

  

  huangyuyang

  committed
  96159de

Commits on Nov 5, 2023

• Fixes ztxz16#360 add stop_token_ids

  Show description for 171e3c1
  wangyumu

  committed
  171e3c1

Commits on Nov 3, 2023

• add support macos

  purpleroc

  committed
  d749e38

Commits on Oct 17, 2023

• 增加释放权重所占内存空间的接口

  siemonchan

  committed
  8742a07

Commits on Oct 13, 2023

• 提升torch2flm转换时4bit模型的量化精度

  siemonchan

  committed
  3397281

Commits on Oct 11, 2023

• 增加baichuan2转换脚本

  

  siemon

  committed
  bfa14e1

Commits on Oct 9, 2023

• fix user_role

  

  huangyuyang

  committed
  ccd0ed6

• 修复transformers>=4.33.0时，activate_adapter作为方法出现在model中的问题

  

  siemon

  committed
  c955c04

Commits on Oct 7, 2023

• Merge pull request ztxz16#338 from TylunasLi/bug_fix_att_opt

  Show description for baf2404
  ztxz16

  authored
  baf2404

Commits on Sep 28, 2023

• ChatGLMModel支持最初的chatglm-6b版本(ztxz16#19)，并减少使用magic number。

  

  cgli

  committed
  7c8d117

Commits on Sep 27, 2023

• Merge pull request ztxz16#331 from fluxlinkage/master

  Show description for 80a3917
  ztxz16

  authored
  80a3917

Commits on Sep 26, 2023

• GLM基础模型可以使用sentencepiece分词器以获得和原版100%相同的输出。

  Show description for 06987b2
  fluxlinkage

  committed
  06987b2

Commits on Sep 25, 2023

• 修复引号错误。

  fluxlinkage

  committed
  5f01e92

Commits on Sep 24, 2023

• optimize pytools tokenizer_*

  lockmatrix

  committed
  e8b1313

• add stream_response_raw

  lockmatrix

  committed
  800885e

Commits on Sep 20, 2023

• 增加GLM基座模型支持。

  fluxlinkage

  committed
  1274cc4

Commits on Aug 18, 2023

• 将fastllm读取模型peft config的方式改为从dict中读取

  

  siemon

  committed
  8a76427

Commits on Aug 16, 2023

• 初步支持huggingface peft

  

  siemon

  committed
  ec31a68

• 修复QWenBase模型generate时停不下来的问题；

  Show description for d27a776
  

  siemon

  committed
  d27a776

Commits on Aug 15, 2023

• fix 模型转换

  

  huangyuyang

  committed
  f4a3547

Commits on Aug 14, 2023

• 支持千问的两行代码加速

  

  siemon

  committed
  ed3fb8e

• fix

  

  huangyuyang

  committed
  4ba5fd5

• support QWen

  

  siemon

  committed
  45c5010

Commits on Aug 12, 2023

• 对齐百川模型中的一些细节

  

  huangyuyang

  committed
  6e211bc

Commits on Aug 9, 2023

• 增加output_logits接口直接输出logits （测试中）

  

  huangyuyang

  committed
  3f12eab

Commits on Aug 4, 2023

• 修复moss模型跑不了的bug

  

  huangyuyang

  committed
  4270d96

Commits on Aug 2, 2023

• fix

  

  huangyuyang

  committed
  2bba479

• Merge pull request ztxz16#186 from denghongcai/feat-optimize-web-demo

  Show description for 5964fdd
  ztxz16

  authored
  5964fdd

• Merge pull request ztxz16#200 from acupofespresso/patch-1

  Show description for eba66e0
  ztxz16

  authored
  eba66e0

• 更新c++的tokenizer

  

  huangyuyang

  committed
  8725348

Commits on Jul 31, 2023

• Update baichuan2flm.py

  Show description for f88563a
  acupofespresso

  authored
  f88563a

Commits on Jul 28, 2023

• feat: 🎸 optimize web demo

  denghongcai

  committed
  95e132d

• 修改直接导出int4的量化方式，提升精度

  

  huangyuyang

  committed
  9b40b4f

Pagination

Previous

Next