Tags

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

b1717

82d6eab2 · main-cmake-pkg : fix build issue (#4665) · Dec 29, 2023
b1716

afd997ab · llama.swiftui : fix infinite loop, ouput timings, buff UI (#4674) · Dec 29, 2023
b1715

c8255f8a · scripts : print list of sync commits · Dec 29, 2023
b1713

38b3de46 · sync : ggml · Dec 29, 2023
b1710

65e5f6da · Fix OpenAI server sampling w.r.t. temp and seed (#4668) · Dec 28, 2023
b1709

ea5497df · gpt2 : Add gpt2 architecture integration (#4555) · Dec 28, 2023
b1708

f6793491 · llama : add AWQ for llama, llama2, mpt, and mistral models (#4593) · Dec 27, 2023
b1707

879b690a · finetune : fix output formatting in print_params (#4653) · Dec 27, 2023
b1705

951010fa · ggml : fix dot product for ARM (#4630) · Dec 27, 2023
b1703

dc68f005 · cuda : fix vmm pool with multi GPU (#4620) · Dec 26, 2023
b1702

de8e4964 · Update comment for AdamW implementation reference. (#4604) · Dec 26, 2023
b1701

77465dad · Fix new CUDA10 compilation errors (#4635) · Dec 26, 2023
b1698

753be377 · llama : add PLaMo model (#3557) · Dec 24, 2023
b1697

5bf3953d · cuda : improve cuda pool efficiency using virtual memory (#4606) · Dec 24, 2023
b1696

708e179e · fallback to CPU buffer if host buffer alloc fails (#4610) · Dec 23, 2023
b1695

925e5584 · ci(docker): fix tags in "Build and push docker image (tagged)" (#4603) · Dec 23, 2023
b1694

61239799 · server : allow to specify custom prompt for penalty calculation (#3727) · Dec 23, 2023
b1693

b9ec82d2 · grammar : check the full vocab only if necessary (opt) (#4306) · Dec 23, 2023
b1692

e0a40022 · CUDA: fixed row rounding for 0 tensor splits (#4594) · Dec 23, 2023
b1691

7082d24c · lookup : add prompt lookup decoding example (#4484) · Dec 22, 2023

Prev
1
…
38
39
40
41
42
43
44
45
46
…
99
Next