nwpie
1 Followers · 0 Following
  • Joined on 2024-06-23
Repositories
3
 Projects Packages Public Activity Starred Repositories
nwpie synced commits to refs/pull/4311/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
417d4f800d Merge 186d4f550373aaf41c398b9c67b878feea5c009c into a38b884c6c
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
Compare 2 commits »
nwpie synced commits to refs/pull/6035/head at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
a13501399d [CANN] Add Ascend NPU backend
f09b7cb609 rm get_work_group_size() by local cache for performance (#8286)
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
Compare 583 commits »
nwpie synced commits to refs/pull/6035/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
4520e91b5b Merge a13501399df068a67a83c98f7a340f38390ed2d0 into 61ecafa390
61ecafa390 passkey : add short intro to README.md [no-ci] (#8317)
aa5898dc53 llama : prefer n_ over num_ prefix (#8308)
6c05752c50 contributing : update guidelines (#8316)
a9554e20b6 [SYCL] Fix WARP_SIZE=16 bug of Intel GPU (#8266)
Compare 545 commits »
nwpie synced commits to refs/pull/6840/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
59e86d19ec Merge 2dd5d1f4b3 into a38b884c6c
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
51d2ebadbb build: Export hf-to-gguf as snakecase
Compare 13 commits »
nwpie synced commits to refs/pull/7187/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
2a9646e6a0 Merge ffca1bb5fbbae893ca9fb8b6262036bf582eab32 into a38b884c6c
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
Compare 4 commits »
nwpie synced commits to refs/pull/7531/head at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
9b38f8bf65 Merge branch 'master' into compilade/refactor-kv-cache
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
51d2ebadbb build: Export hf-to-gguf as snakecase
1e920018d3 doc: Add context for why we add an explicit pytorch source
Compare 39 commits »
nwpie synced commits to refs/pull/7531/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
20b7a47ad6 Merge 9b38f8bf65 into a9554e20b6
a9554e20b6 [SYCL] Fix WARP_SIZE=16 bug of Intel GPU (#8266)
e235b267a2 py : switch to snake_case (#8305)
f09b7cb609 rm get_work_group_size() by local cache for performance (#8286)
9b38f8bf65 Merge branch 'master' into compilade/refactor-kv-cache
Compare 38 commits »
nwpie synced commits to refs/pull/7705/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
7c7b9c0bae Merge 5175117a09 into f09b7cb609
f09b7cb609 rm get_work_group_size() by local cache for performance (#8286)
Compare 2 commits »
nwpie synced commits to refs/pull/7844/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
20c683d11e Merge 181c0e3b0f into a38b884c6c
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
51d2ebadbb build: Export hf-to-gguf as snakecase
Compare 12 commits »
nwpie synced commits to refs/pull/8035/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
f9b0e825c9 Merge c0ba35740cd9d75ffe841d0855fc0a93e1d1359f into a38b884c6c
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
51d2ebadbb build: Export hf-to-gguf as snakecase
Compare 13 commits »
nwpie synced commits to refs/pull/8039/head at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
906476f6ff style: spaces
4db8c0d5d9 Update bruteforce test: add more models
11ac641c1e Update bruteforce test: header files location
2f150197e4 Better leading space removal
8f5e1e0c76 'viking' detokenizer clean spaces
Compare 114 commits »
nwpie synced commits to refs/pull/8039/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
bde926acc1 Merge 906476f6ff into f09b7cb609
f09b7cb609 rm get_work_group_size() by local cache for performance (#8286)
906476f6ff style: spaces
4db8c0d5d9 Update bruteforce test: add more models
11ac641c1e Update bruteforce test: header files location
Compare 89 commits »
nwpie synced commits to refs/pull/8048/head at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
f324f42c4e gguf-hash: missing stdalign.h in windows bypass
d8dd43f94f gguf-hash: add cpp and python implementation of layer + model wide hashing
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
Compare 60 commits »
nwpie synced commits to refs/pull/8048/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
cde69b70ac Merge f324f42c4e into a38b884c6c
f324f42c4e gguf-hash: missing stdalign.h in windows bypass
d8dd43f94f gguf-hash: add cpp and python implementation of layer + model wide hashing
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
Compare 6 commits »
nwpie synced commits to refs/pull/8087/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:33 +03:00
f484b12eec Merge 9fa007c850 into a38b884c6c
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
51d2ebadbb build: Export hf-to-gguf as snakecase
Compare 15 commits »
nwpie synced commits to compilade/refactor-kv-cache at nwpie/llama.cpp from mirror 2024-07-05 09:23:32 +03:00
9b38f8bf65 Merge branch 'master' into compilade/refactor-kv-cache
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
51d2ebadbb build: Export hf-to-gguf as snakecase
1e920018d3 doc: Add context for why we add an explicit pytorch source
Compare 39 commits »
nwpie synced commits to gg/fix-python-names at nwpie/llama.cpp from mirror 2024-07-05 09:23:32 +03:00
91deef4606 py : rename requirements for convert_legacy_llama.py
902de8826b gguf-py : use snake_case in scripts entrypoint export
3e3cc7102f cont : fix link
c172b322c2 cont
Compare 4 commits »
nwpie synced commits to gg/indent-load at nwpie/llama.cpp from mirror 2024-07-05 09:23:32 +03:00
dc91715b44 llama : minor indentation during tensor loading
aa5898dc53 llama : prefer n_ over num_ prefix (#8308)
6c05752c50 contributing : update guidelines (#8316)
a9554e20b6 [SYCL] Fix WARP_SIZE=16 bug of Intel GPU (#8266)
e235b267a2 py : switch to snake_case (#8305)
Compare 7 commits »
nwpie synced commits to master at nwpie/llama.cpp from mirror 2024-07-05 09:23:32 +03:00
61ecafa390 passkey : add short intro to README.md [no-ci] (#8317)
aa5898dc53 llama : prefer n_ over num_ prefix (#8308)
6c05752c50 contributing : update guidelines (#8316)
a9554e20b6 [SYCL] Fix WARP_SIZE=16 bug of Intel GPU (#8266)
e235b267a2 py : switch to snake_case (#8305)
Compare 6 commits »
nwpie synced commits to refs/pull/4012/merge at nwpie/llama.cpp from mirror 2024-07-05 09:23:32 +03:00
52ab632416 Merge 4e23f8a81b into a38b884c6c
a38b884c6c cli: add EOT when user hit Ctrl+C (#8296)
d7fd29fff1 llama : add OpenELM support (#7359)
6f63d646c1 tokenize : add --show-count (token) option (#8299)
51d2ebadbb build: Export hf-to-gguf as snakecase
Compare 34 commits »