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

Removed unnecessary method - LlamaModel.permute is used instead.

Browse source
This commit is contained in:
Stanisław Szymczyk 2024-05-09 19:41:39 +02:00
parent 4ebb52cfc2
commit 9acc3ecf34
1 changed files with 0 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

8
convert-hf-to-gguf.py
View file

@ -2370,14 +2370,6 @@ class ArcticModel(Model):
self.gguf_writer.add_vocab_size(hparams["vocab_size"])
self.gguf_writer.add_rope_dimension_count(hparams["hidden_size"] // hparams["num_attention_heads"])
@staticmethod
def permute(weights: Tensor, n_head: int, n_head_kv: int | None):
if n_head_kv is not None and n_head != n_head_kv:
n_head = n_head_kv
return (weights.reshape(n_head, 2, weights.shape[0] // n_head // 2, *weights.shape[1:])
.swapaxes(1, 2)
.reshape(weights.shape))
_experts: list[dict[str, Tensor]] | None = None
def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]: