tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-04 18:16:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

convert : update phi-2 to latest HF repo (#4903)

Browse Source 
* convert : update phi-2 to latest HF repo

ggml-ci

* py : try to fix flake stuff
This commit is contained in:
Georgi Gerganov
2024-01-13 13:44:37 +02:00
committed by GitHub
parent de473f5f8e
commit 15ebe59210
4 changed files with 65 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
gguf-py/gguf/constants.py
View File

@ -389,6 +389,9 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_QKV,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_DOWN,

2
gguf-py/gguf/tensor_mapping.py
View File

@ -191,6 +191,7 @@ class TensorNameMap:
"transformer.h.{bid}.mlp.w1", # qwen
"h.{bid}.mlp.c_fc", # gpt2
"transformer.h.{bid}.mlp.fc1", # phi2
"model.layers.{bid}.mlp.fc1", # phi2
"model.layers.layers.{bid}.mlp.up_proj", # plamo
),
@ -232,6 +233,7 @@ class TensorNameMap:
"model.layers.{bid}.mlp.dense_4h_to_h", # persimmon
"h.{bid}.mlp.c_proj", # gpt2
"transformer.h.{bid}.mlp.fc2", # phi2
"model.layers.{bid}.mlp.fc2", # phi2
"model.layers.layers.{bid}.mlp.down_proj", # plamo
),