vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : add support for Orion-14B (#5118)

Browse source 
* add support for Orion-14B(https://huggingface.co/OrionStarAI/Orion-14B-Chat)

* flake8 support

* Update llama.cpp

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* Update llama.cpp

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* Update llama.cpp

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* Update llama.cpp

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* Update llama.cpp

Co-authored-by: slaren <slarengh@gmail.com>

* Update llama.cpp

* Update llama.cpp

---------

Co-authored-by: lixiaopu <lixiaopu@cmcm.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: slaren <slarengh@gmail.com>
This commit is contained in:
sharpHL 2024-01-28 16:00:30 +08:00 • committed by GitHub
parent 39baaf55a1
commit f2e69d28c0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 291 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

24
gguf-py/gguf/constants.py
View file

@ -101,6 +101,7 @@ class MODEL_ARCH(IntEnum):
PHI2 = auto()
PLAMO = auto()
CODESHELL = auto()
ORION = auto()
class MODEL_TENSOR(IntEnum):
@ -151,6 +152,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
MODEL_ARCH.PHI2: "phi2",
MODEL_ARCH.PLAMO: "plamo",
MODEL_ARCH.CODESHELL: "codeshell",
MODEL_ARCH.ORION: "orion",
}
TENSOR_NAMES: dict[MODEL_TENSOR, str] = {
@ -427,7 +429,23 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
]
],
MODEL_ARCH.ORION: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ROPE_FREQS,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.ATTN_ROT_EMBD,
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_GATE,
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
],
# TODO
}
@ -452,6 +470,10 @@ MODEL_TENSOR_SKIP: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.ROPE_FREQS,
MODEL_TENSOR.ATTN_ROT_EMBD,
],
MODEL_ARCH.ORION: [
MODEL_TENSOR.ROPE_FREQS,
MODEL_TENSOR.ATTN_ROT_EMBD,
],
}
#