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

Update convert_hf_to_gguf.py

Browse source 
Co-authored-by: compilade <git@compilade.net>
This commit is contained in:
Jeffrey Morgan 2024-07-27 00:41:30 -07:00 • committed by GitHub
parent e6d5bed7d3
commit 658041d107
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
convert_hf_to_gguf.py
View file

@ -1573,7 +1573,7 @@ class LlamaModel(Model):
if rope_scaling := self.find_hparam(["rope_scaling"], optional=True):
if rope_scaling.get("rope_type", '').lower() == "llama3":
base = self.hparams.get("rope_theta", 10000.0)
dim = int((self.hparams["hidden_size"] // self.hparams["num_attention_heads"]) * self.hparams.get("partial_rotary_embeddings", 1.0))
dim = self.hparams["hidden_size"] // self.hparams["num_attention_heads"]
freqs = 1.0 / (base ** (torch.arange(0, dim, 2, dtype=torch.float32) / dim))
factor = rope_scaling.get("factor", 8.0)