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

convert: remove most n_mult usage

Browse source
This commit is contained in:
Green Sky 2023-09-09 17:43:57 +02:00
parent 21ac3a1503
commit ecd7beddf0
No known key found for this signature in database
1 changed files with 0 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

6
convert.py
View file

@ -145,7 +145,6 @@ GGML_FILE_TYPE_TO_DATA_TYPE: dict[GGMLFileType, DataType] = {
class Params:
n_vocab: int
n_embd: int
n_mult: int
n_layer: int
n_ctx: int
n_ff: int
@ -197,7 +196,6 @@ class Params:
return Params(
n_vocab = n_vocab,
n_embd = n_embd,
n_mult = n_mult,
n_layer = n_layer,
n_ctx = -1,
n_ff = n_ff,
@ -225,8 +223,6 @@ class Params:
else:
f_rope_scale = None
n_mult = Params.find_n_mult(n_ff, n_embd)
if "max_sequence_length" in config:
n_ctx = config["max_sequence_length"]
elif "max_position_embeddings" in config:
@ -238,7 +234,6 @@ class Params:
return Params(
n_vocab = n_vocab,
n_embd = n_embd,
n_mult = n_mult,
n_layer = n_layer,
n_ctx = n_ctx,
n_ff = n_ff,
@ -285,7 +280,6 @@ class Params:
return Params(
n_vocab = n_vocab,
n_embd = n_embd,
n_mult = n_mult,
n_layer = n_layer,
n_ctx = n_ctx,
n_ff = n_ff,