diff options
Diffstat (limited to 'gguf-py')
| -rw-r--r-- | gguf-py/gguf/constants.py | 3 | ||||
| -rw-r--r-- | gguf-py/gguf/tensor_mapping.py | 2 |
2 files changed, 5 insertions, 0 deletions
diff --git a/gguf-py/gguf/constants.py b/gguf-py/gguf/constants.py index f0a1c51f..972b4e9a 100644 --- a/gguf-py/gguf/constants.py +++ b/gguf-py/gguf/constants.py @@ -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, diff --git a/gguf-py/gguf/tensor_mapping.py b/gguf-py/gguf/tensor_mapping.py index 24a08903..e5b14610 100644 --- a/gguf-py/gguf/tensor_mapping.py +++ b/gguf-py/gguf/tensor_mapping.py @@ -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 ), |