Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-06 02:48:57 +01:00
Code Issues Packages Projects Releases Wiki Activity

gguf : single pass for writing tensors + refactoring writer

Browse Source
This commit is contained in:
M. Yusuf Sarıgöz 2023-08-17 16:48:49 +03:00
parent f31e9230ad
commit dce07c3121
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
convert-llama-h5-to-gguf.py
View File

@ -267,7 +267,7 @@ for part_name in part_names:
data = data.astype(data_dtype) data = data.astype(data_dtype)
print(name + ", n_dims = " + n_dims + ", " + str(old_dtype) + " --> " + str(data.dtype)) print(name + ", n_dims = " + str(n_dims) + ", " + str(old_dtype) + " --> " + str(data.dtype))
gguf_writer.add_tensor(name, data) gguf_writer.add_tensor(name, data)