haraldwolff
/
whisper.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

pull/1812/head
Georgi Gerganov 2024-01-27 17:24:53 +02:00
parent 7fe3ed5e00
commit ef3c9ed9eb
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
3 changed files with 832 additions and 276 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1096
examples/talk-llama/llama.cpp
View File

File diff suppressed because it is too large Load Diff

9
examples/talk-llama/llama.h
View File

@ -107,6 +107,7 @@ extern "C" {
LLAMA_FTYPE_MOSTLY_IQ2_XXS = 19, // except 1d tensors
LLAMA_FTYPE_MOSTLY_IQ2_XS = 20, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q2_K_S = 21, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q3_K_XS = 22, // except 1d tensors
LLAMA_FTYPE_GUESSED = 1024, // not specified in the model file
};
@ -774,6 +775,14 @@ extern "C" {
float p,
size_t min_keep);
/// @details Dynamic temperature implementation described in the paper https://arxiv.org/abs/2309.02772.
LLAMA_API void llama_sample_entropy(
struct llama_context * ctx,
llama_token_data_array * candidates_p,
float min_temp,
float max_temp,
float exponent_val);
LLAMA_API void llama_sample_temp(
struct llama_context * ctx,
llama_token_data_array * candidates,

3
examples/talk-llama/unicode.h
View File

@ -2,8 +2,9 @@
#include <cassert>
#include <stdexcept>
#include <vector>
#include <string>
#include <unordered_map>
#include <vector>
static const std::vector<std::pair<uint32_t, uint32_t>> digit_ranges = {
{0x30, 0x39}, {0xB2, 0xB3}, {0xB9, 0xB9}, {0x660, 0x669}, {0x6F0, 0x6F9}, {0x7C0, 0x7C9}, {0x966, 0x96F}, {0x9E6, 0x9EF}, {0xA66, 0xA6F}, {0xAE6, 0xAEF}, {0xB66, 0xB6F}, {0xBE6, 0xBEF}, {0xC66, 0xC6F},