[Quant][Kernel] Remove GPTQ Group/Dynamic Activation Ordering (#54809)
Signed-off-by: Roderick-Wu <roderickwu2003@gmail.com> Signed-off-by: Roderick Wu <roderickwu2003@gmail.com> Signed-off-by: mgoin <mgoin64@gmail.com> Co-authored-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
@@ -22,12 +22,10 @@ os.environ["TOKENIZERS_PARALLELISM"] = "true"
|
||||
MAX_MODEL_LEN = 1024
|
||||
|
||||
MODELS = [
|
||||
# act_order==True, group_size=128
|
||||
("TheBloke/TinyLlama-1.1B-Chat-v1.0-GPTQ", "main"),
|
||||
# 8-bit, act_order==True, group_size=channelwise
|
||||
("LnL-AI/TinyLlama-1.1B-Chat-v1.0-GPTQ-4bit", "main"),
|
||||
# desc_act=True is a no-op for channelwise quantization.
|
||||
("TheBloke/TinyLlama-1.1B-Chat-v1.0-GPTQ", "gptq-8bit--1g-actorder_True"),
|
||||
# 4-bit, act_order==True, group_size=128
|
||||
("TechxGenus/gemma-1.1-2b-it-GPTQ", "main"),
|
||||
("TechxGenus/gemma-2b-it-GPTQ", "main"),
|
||||
]
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user