diff options
Diffstat (limited to 'gguf-py/gguf/tensor_mapping.py')
-rw-r--r-- | gguf-py/gguf/tensor_mapping.py | 4 |
1 files changed, 4 insertions, 0 deletions
diff --git a/gguf-py/gguf/tensor_mapping.py b/gguf-py/gguf/tensor_mapping.py index e5750d41..990fe63c 100644 --- a/gguf-py/gguf/tensor_mapping.py +++ b/gguf-py/gguf/tensor_mapping.py @@ -137,6 +137,7 @@ class TensorNameMap: "layers.{bid}.attention.wk", # llama-pth "encoder.layer.{bid}.attention.self.key", # bert "transformer.h.{bid}.attn.k_proj", # gpt-j + "transformer.h.{bid}.attn.k", # refact "model.layers.layers.{bid}.self_attn.k_proj", # plamo "model.layers.{bid}.attention.wk", # internlm2 "transformer.decoder_layer.{bid}.multi_head_attention.key" # Grok @@ -148,6 +149,7 @@ class TensorNameMap: "layers.{bid}.attention.wv", # llama-pth "encoder.layer.{bid}.attention.self.value", # bert "transformer.h.{bid}.attn.v_proj", # gpt-j + "transformer.h.{bid}.attn.v", # refact "model.layers.layers.{bid}.self_attn.v_proj", # plamo "model.layers.{bid}.attention.wv", # internlm2 "transformer.decoder_layer.{bid}.multi_head_attention.value" # Grok @@ -229,6 +231,7 @@ class TensorNameMap: "layers.{bid}.feed_forward.w3", # llama-pth "encoder.layer.{bid}.intermediate.dense", # bert "transformer.h.{bid}.mlp.fc_in", # gpt-j + "transformer.h.{bid}.mlp.linear_3", # refact "language_model.encoder.layers.{bid}.mlp.dense_h_to_4h", # persimmon "model.layers.{bid}.mlp.dense_h_to_4h", # persimmon "transformer.h.{bid}.mlp.w1", # qwen @@ -266,6 +269,7 @@ class TensorNameMap: "model.layers.layers.{bid}.mlp.gate_proj", # plamo "model.layers.{bid}.feed_forward.w1", # internlm2 "encoder.layers.{bid}.mlp.fc12", # nomic-bert + "transformer.h.{bid}.mlp.linear_1", # refact ), MODEL_TENSOR.FFN_GATE_EXP: ( |