llama.cpp
ggml : full ALiBi support
#7192
Merged
Overview
Commits
10
Changes
View On GitHub
Commits
  • ggml : full ALiBi support
    ggerganov committed 2 years ago
  • ggml : update ggml_soft_max_ext() CUDA, SYCL
    ggerganov committed 2 years ago
  • ggml : ggml_flash_attn_ext() support ALiBi (CPU)
    ggerganov committed 2 years ago
  • ggml : ggml_flash_attn_ext() support ALiBi (Metal)
    ggerganov committed 2 years ago
  • ggml : fix warning
    ggerganov committed 2 years ago
  • ggml : ggml_flash_attn_ext() support ALiBi (CUDA)
    ggerganov committed 2 years ago
  • ggml : fix assert message
    ggerganov committed 2 years ago
  • vulkan : add dev notes
    ggerganov committed 2 years ago
  • ggml : require mask when using ALiBi
    ggerganov committed 2 years ago
  • convert : fix convert for refact models
    ggerganov committed 2 years ago
FAQ Terms Privacy Refunds Impressum
Loading