 EyeDeck
|
39f3fec913
Broaden GPTQ-for-LLaMA branch support (#820)
|
2 gadi atpakaļ |
 oobabooga
|
3d6cb5ed63
Minor rewrite
|
2 gadi atpakaļ |
|
oobabooga
|
f3a2e0b8a9
Disable pre_layer when the model type is not llama
|
2 gadi atpakaļ |
 catalpaaa
|
4ab679480e
allow quantized model to be loaded from model dir (#760)
|
2 gadi atpakaļ |
 OWKenobi
|
ee4547cd34
Detect "vicuna" as llama model type (#772)
|
2 gadi atpakaļ |
|
oobabooga
|
1cb9246160
Adapt to the new model names
|
2 gadi atpakaļ |
|
oobabooga
|
010b259dde
Update documentation
|
2 gadi atpakaļ |
|
oobabooga
|
0bec15ebcd
Reorder imports
|
2 gadi atpakaļ |
 Maya Eary
|
41ec682834
Disable kernel threshold for gpt-j
|
2 gadi atpakaļ |
|
Maya Eary
|
1c075d8d21
Fix typo
|
2 gadi atpakaļ |
|
Maya Eary
|
c8207d474f
Generalized load_quantized
|
2 gadi atpakaļ |
|
oobabooga
|
49c10c5570
Add support for the latest GPTQ models with group-size (#530)
|
2 gadi atpakaļ |
|
EyeDeck
|
dcfd866402
Allow loading of .safetensors through GPTQ-for-LLaMa
|
2 gadi atpakaļ |
|
oobabooga
|
db4219a340
Update comments
|
2 gadi atpakaļ |
|
oobabooga
|
7618f3fe8c
Add -gptq-preload for 4-bit offloading (#460)
|
2 gadi atpakaļ |
|
oobabooga
|
9a3bed50c3
Attempt at fixing 4-bit with CPU offload
|
2 gadi atpakaļ |
 askmyteapot
|
53b6a66beb
Update GPTQ_Loader.py
|
2 gadi atpakaļ |
|
oobabooga
|
265ba384b7
Rename a file, add deprecation warning for --load-in-4bit
|
2 gadi atpakaļ |
