 anzz1
|
34ab526843
(Windows) Set console to UTF-8 on init (#420)
|
2 лет назад |
 Georgi Gerganov
|
c2b25b6912
Fix colors enabling on WIN32
|
2 лет назад |
|
Georgi Gerganov
|
79b2b266db
If n_predict == -1, generate forever
|
2 лет назад |
|
Georgi Gerganov
|
e2d490dafd
Inifinite generation via context swapping (#71)
|
2 лет назад |
|
Georgi Gerganov
|
03f7e33560
Cleanup STL headers + fix embedding examples + minor stuff
|
2 лет назад |
|
Georgi Gerganov
|
55ad42af84
Move chat scripts into "./examples"
|
2 лет назад |
 slaren
|
459e93cce0
Add AVX2 implementation of dequantize_row_q4_1 (#505)
|
2 лет назад |
|
Georgi Gerganov
|
a316a425d0
Overhaul the examples structure
|
2 лет назад |
|
Georgi Gerganov
|
ecbe466a36
Retire the ggml_mul_mat() branch for transposed src0 (#500)
|
2 лет назад |
|
Georgi Gerganov
|
502a400192
Disable prompt verbosity by default and add option to enable (#480)
|
2 лет назад |
|
slaren
|
09aecbf628
Add AVX2 implementation of dequantize_row_q4_0 (#467)
|
2 лет назад |
|
Georgi Gerganov
|
4640eff23d
Don't interefe with BLAS for large prompts by running only 1 thread
|
2 лет назад |
|
Georgi Gerganov
|
ab77d76312
Add longer DAN prompt for testing big batch numbers
|
2 лет назад |
|
slaren
|
29b7baab67
Add timings for the prompt evaluation (#478)
|
2 лет назад |
|
Georgi Gerganov
|
4a7129acd2
Remove obsolete information from README
|
2 лет назад |
|
Georgi Gerganov
|
6b6dbc8910
Remove obsolete assert and fix compiler warning
|
2 лет назад |
|
Georgi Gerganov
|
2a2e63ce05
Fix nasty bug in ggml_compute_forward_mul_mat_f32() and reenable BLAS
|
2 лет назад |
|
anzz1
|
e899bf54b2
bounds checking for input prefix (#492)
|
2 лет назад |
|
anzz1
|
fbd4d38c64
feat: '--in-prefix STRING' option (#426)
|
2 лет назад |
 Jed Fox
|
58e6c9f36f
Add support for file load progress reporting callbacks (#434)
|
2 лет назад |
 Doomsdayrs
|
36d07532ef
Add missing struct annotation (#483)
|
2 лет назад |
 Chris Kuehl
|
6f1ee4b640
Fix crash for 65B model with pre-allocated memory (#485)
|
2 лет назад |
|
Georgi Gerganov
|
8520fc310e
Disable BLAS altogether - the bug is not just for qunatized mat mul
|
2 лет назад |
|
Georgi Gerganov
|
b3f460e941
Disable BLAS branch in mul_mat - seems there is a bug
|
2 лет назад |
|
Georgi Gerganov
|
04c6f5ed6f
Immediately start processing the prompt before user input has been provided (#476)
|
2 лет назад |
|
Georgi Gerganov
|
7a9b6c3a8b
Reduce memory usage and allocate enough memory for largest context (#473)
|
2 лет назад |
|
Georgi Gerganov
|
31572d9665
Temporary bump the memory buffer size - hopefully fix issues from 483bab2e
|
2 лет назад |
 Gary Mulder
|
f4f5362edb
Update README.md (#444)
|
2 лет назад |
 rabidcopy
|
863f65e2e3
fix instruct mode (#445)
|
2 лет назад |
|
Georgi Gerganov
|
afd220d9c6
Properly free llama_context on failure
|
2 лет назад |
