mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 30 Packages Projects Releases Wiki Activity

models : fix YaRN regression + consolidate logic (#18006)

Browse Source 
* models : fix YaRN regression + consolidate logic

* cont : fix the fix

* cont : remove header

* cont : add header
This commit is contained in:
Georgi Gerganov
2025-12-14 08:34:56 +02:00
committed by GitHub
parent a63cbafbbc
commit 609a2d0268
6 changed files with 40 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-kv-cache.cpp
+1 -1
View File
@@ -1372,7 +1372,7 @@ ggml_tensor * llama_kv_cache::build_rope_shift(
const auto & yarn_ext_factor = cparams.yarn_ext_factor;
const auto & yarn_beta_fast = cparams.yarn_beta_fast;
const auto & yarn_beta_slow = cparams.yarn_beta_slow;
const auto & yarn_attn_factor = llama_hparams::yarn_attn_factor_adjust(cparams.yarn_attn_factor, cparams.rope_freq_scale, cparams.yarn_ext_factor);
const auto & yarn_attn_factor = cparams.yarn_attn_factor;
const auto & n_rot = hparams.n_rot;
const auto & rope_type = hparams.rope_type == LLAMA_ROPE_TYPE_MROPE || hparams.rope_type == LLAMA_ROPE_TYPE_IMROPE