From 9acc3ecf34ca7ce579965876e57ed62201c2b95a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Stanis=C5=82aw=20Szymczyk?= Date: Thu, 9 May 2024 19:41:39 +0200 Subject: [PATCH] Removed unnecessary method - LlamaModel.permute is used instead. --- convert-hf-to-gguf.py | 8 -------- 1 file changed, 8 deletions(-) diff --git a/convert-hf-to-gguf.py b/convert-hf-to-gguf.py index 3c415a604..170dea060 100755 --- a/convert-hf-to-gguf.py +++ b/convert-hf-to-gguf.py @@ -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]]: