CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

YaRN : store rope scaling type as int32_t in memory (#5285)

Browse Source 
* YaRN : store rope scaling type as int32_t in memory

* llama : store mapped names as const char *
This commit is contained in:
Jared Van Bortel
2024-02-03 06:22:06 -05:00
committed by GitHub
parent 6a66c5071a
commit 1ec3332ade
3 changed files with 14 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
common/common.h
View File

@@ -75,8 +75,7 @@ struct gpt_params {
float yarn_beta_fast = 32.0f; // YaRN low correction dim
float yarn_beta_slow = 1.0f; // YaRN high correction dim
int32_t yarn_orig_ctx = 0; // YaRN original context length
int8_t rope_scaling_type = LLAMA_ROPE_SCALING_UNSPECIFIED; // TODO: better to be int32_t for alignment
// pinging @cebtenzzre
int32_t rope_scaling_type = LLAMA_ROPE_SCALING_UNSPECIFIED;
// // sampling parameters
struct llama_sampling_params sparams;