 Georgi Gerganov
|
c1c7026b47
Fix python stuff (#109)
|
2 gadi atpakaļ |
 qunash
|
467b149761
Refactoring `convert-pth-to-ggml.py`: more concise and readable (#109)
|
2 gadi atpakaļ |
 Bernat Vadell
|
2af23d3043
🚀 Dockerize llamacpp (#132)
|
2 gadi atpakaļ |
 Ronsor
|
956dfda8ad
Use `tokenizer.vocab_size()` instead of hardcoding 32000 in convert-pth-to-ggml.py (#142)
|
2 gadi atpakaļ |
 Val Kharitonov
|
2a20f48efa
Fix UTF-8 handling (including colors) (#79)
|
2 gadi atpakaļ |
|
Georgi Gerganov
|
7c9e54e55e
Revert "weights_only" arg - this causing more trouble than help
|
2 gadi atpakaļ |
 Oleksandr Nikitin
|
b9bd1d0141
python/pytorch compat notes (#44)
|
2 gadi atpakaļ |
 deepdiffuser
|
a93120236f
use weights_only in conversion script (#32)
|
2 gadi atpakaļ |
|
Georgi Gerganov
|
007a8f6f45
Support all LLaMA models + change Q4_0 quantization storage
|
2 gadi atpakaļ |
|
Georgi Gerganov
|
70bc0b8b15
Fix a bug in the rope calculation
|
2 gadi atpakaļ |
|
Georgi Gerganov
|
26c0846629
Initial release
|
2 gadi atpakaļ |
