Tags

Tags give the ability to mark specific points in history as being important

b3003

d298382a · main: replace --no-special with --special (#7534) · May 27, 2024
b3001

c429b33b · llama : add Smaug 70B support (#7402) · May 26, 2024
b2998

9588f196 · train : change default FA argument (#7528) · May 25, 2024
b2996

00c63907 · main : don't print special tokens with --grammar (#6923) · May 25, 2024
b2995

faa0e697 · ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433) · May 25, 2024
b2994

9791f402 · android : module (#7502) · May 25, 2024
b2993

902184dd · fix missing slash in `fs_get_cache_directory()` (#7503) · May 25, 2024
b2992

57684331 · Make tokenize CLI tool have nicer command line arguments. (#6188) · May 25, 2024
b2989

27891f6d · docker.yml: disable light-intel and server-intel test (#7515) · May 24, 2024
b2988

fbca2f27 · Add support for ArcticForCausalLM (#7020) · May 24, 2024
b2985

1debe727 · ggml : silence UB sanitizer error during iq2_xxs quantization (#0) · May 23, 2024
b2984

007489e8 · Fix phi3 chat template confusion with zephyr (#7449) · May 23, 2024
b2982

3015851c · llama : add getters for n_threads/n_threads_batch (#7464) · May 23, 2024
b2981

55ac3b7a · ci : use Pythia models instead of OpenLlama (#7470) · May 23, 2024
b2979

9b82476e · Add missing inference support for GPTNeoXForCausalLM (Pythia and GPT-NeoX base models) (#7461) · May 23, 2024
b2978

a61a94e5 · llama : rename n_ctx -> cache.size, less confusing (#0) · May 23, 2024
b2976

d48c88cb · ggml : remove ggml_flash_attn and ggml_flash_ff (#7463) · May 23, 2024
b2974

1b1e27cb · Update vulkan rope implementation to support frequency factors (#7475) · May 23, 2024
b2973

fbf777d2 · main : minor (#7462) · May 23, 2024
b2972

cd93a28c · CUDA: fix FA out-of-bounds reads (#7479) · May 23, 2024

Prev
1
2
3
4
5
6
7
8
…
99
Next