Skip to content
This repository has been archived by the owner on Oct 25, 2024. It is now read-only.

fix int8 skip module config #1682

Merged
merged 7 commits into from
Aug 9, 2024
Merged
Changes from 5 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -831,7 +831,10 @@ def __init__(
self.double_quant_bits = double_quant_bits
self.double_quant_use_sym = double_quant_use_sym
self.double_quant_group_size = double_quant_group_size
self.llm_int8_skip_modules = kwargs.get("llm_int8_skip_modules", ["lm_head", "output_layer", "embed_out"])
# "transformer.output_layer" for chatglm series model.
# "embed_out" for dolly v2 series model.
self.llm_int8_skip_modules = kwargs.get("llm_int8_skip_modules",
["lm_head", "transformer.output_layer", "embed_out"])
self.use_ggml = use_ggml
self.use_quant = use_quant
self.use_neural_speed = use_neural_speed
Expand Down
Loading