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

llama : optimize DeepSeek MLA implementation

Browse source
This commit is contained in:
Stanisław Szymczyk 2025-01-25 18:10:22 +01:00
parent f0ce53f158
commit de538aa329
10 changed files with 96 additions and 41 deletions
Download patch file Download diff file Expand all files Collapse all files

6
gguf-py/gguf/constants.py
View file

@ -356,6 +356,8 @@ class MODEL_TENSOR(IntEnum):
ATTN_Q_B = auto()
ATTN_KV_A_MQA = auto()
ATTN_KV_B = auto()
ATTN_K_B = auto()
ATTN_V_B = auto()
ATTN_Q_A_NORM = auto()
ATTN_KV_A_NORM = auto()
FFN_SUB_NORM = auto()
@ -543,6 +545,8 @@ TENSOR_NAMES: dict[MODEL_TENSOR, str] = {
MODEL_TENSOR.ATTN_Q_B: "blk.{bid}.attn_q_b",
MODEL_TENSOR.ATTN_KV_A_MQA: "blk.{bid}.attn_kv_a_mqa",
MODEL_TENSOR.ATTN_KV_B: "blk.{bid}.attn_kv_b",
MODEL_TENSOR.ATTN_K_B: "blk.{bid}.attn_k_b",
MODEL_TENSOR.ATTN_V_B: "blk.{bid}.attn_v_b",
MODEL_TENSOR.ATTN_Q_A_NORM: "blk.{bid}.attn_q_a_norm",
MODEL_TENSOR.ATTN_KV_A_NORM: "blk.{bid}.attn_kv_a_norm",
MODEL_TENSOR.ATTN_SUB_NORM: "blk.{bid}.attn_sub_norm",
@ -1333,6 +1337,8 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.ATTN_Q_B,
MODEL_TENSOR.ATTN_KV_A_MQA,
MODEL_TENSOR.ATTN_KV_B,
MODEL_TENSOR.ATTN_K_B,
MODEL_TENSOR.ATTN_V_B,
MODEL_TENSOR.ATTN_Q_A_NORM,
MODEL_TENSOR.ATTN_KV_A_NORM,
MODEL_TENSOR.ATTN_OUT,

8
gguf-py/gguf/tensor_mapping.py
View file

@ -586,6 +586,14 @@ class TensorNameMap:
"model.layers.{bid}.self_attn.kv_b_proj", # deepseek2
),
MODEL_TENSOR.ATTN_K_B: (
"model.layers.{bid}.self_attn.k_b_proj", # deepseek2
),
MODEL_TENSOR.ATTN_V_B: (
"model.layers.{bid}.self_attn.v_b_proj", # deepseek2
),
MODEL_TENSOR.ATTN_Q_A_NORM: (
"model.layers.{bid}.self_attn.q_a_layernorm", # deepseek2
),