Code

gguf : deduplicate#2629

Code

Merged

ggerganov merged 9 commits intoggufggerganov/llama.cpp:gguffrom gguf-deduplicateggerganov/llama.cpp:gguf-deduplicateCopy head branch name to clipboardAug 16, 2023

+1,776-7,544

Commits

Commits on Aug 16, 2023

• gguf : better type names

  ggerganov

  committed
  d72a23e

  Copy full SHA for d72a23e

• dedup : CPU + Metal is working

  ggerganov

  committed
  1891c92

  Copy full SHA for 1891c92

• ggml : fix warnings about unused results

  ggerganov

  committed
  dd6eaa3

  Copy full SHA for dd6eaa3

• llama.cpp : fix line feed and compiler warning

  klosax

  authored
  a499313

  Copy full SHA for a499313

• llama : fix strncpy warning + note token_to_str does not write null

  ggerganov

  committed
  5b94b14

  Copy full SHA for 5b94b14

• llama : restore the original load/save session implementation

  Show description for 6412e97
  ggerganov

  committed
  6412e97

  Copy full SHA for 6412e97

• convert-llama-h5-to-gguf.py : support alt ctx param name

  klosax

  authored
  f3e90f2

  Copy full SHA for f3e90f2

• ggml : assert when using ggml_mul with non-F32 src1

  ggerganov

  committed
  c290f3e

  Copy full SHA for c290f3e

• examples : dedup simple

  ggerganov

  committed
  795ec70

  Copy full SHA for 795ec70