Temporarily hard-coded mscale value for DeepSeek-V2 (FIXME!).
This commit is contained in:
parent
7e4786bbfb
commit
71a742256c
2
ggml.c
2
ggml.c
|
|
@ -14073,6 +14073,8 @@ static void rope_yarn(
|
|||
// Get n-d magnitude scaling corrected for interpolation
|
||||
mscale *= 1.0f + 0.1f * logf(1.0f / freq_scale);
|
||||
}
|
||||
// TODO ugly hack for DeepSeek-V2 until we find a solution
|
||||
mscale = 1.0;
|
||||
*cos_theta = cosf(theta) * mscale;
|
||||
*sin_theta = sinf(theta) * mscale;
|
||||
}
|
||||
|
|
|
|||
Loading…
Reference in New Issue