Merge branch 'v0.0.6_develop_sugon' into 'main'

v0.0.6 See merge request dcutoolkit/deeplearing/autoawq_kernels!1

Merge branch 'v0.0.6_develop_sugon' into 'main'
v0.0.6 See merge request dcutoolkit/deeplearing/autoawq_kernels!1
9f217825 · gaoqiong · b2c05ad6 · 1c46b800 · 9f217825 · 9f217825
Commit 9f217825 authored Jul 03, 2024 by gaoqiong
20 changed files
--- a/awq_ext/exllamav2/cuda/quant/qdq_3.cuh
+++ b/awq_ext/exllamav2/cuda/quant/qdq_3.cuh
--- a/awq_ext/exllamav2/cuda/quant/qdq_4.cuh
+++ b/awq_ext/exllamav2/cuda/quant/qdq_4.cuh
--- a/awq_ext/exllamav2/cuda/quant/qdq_5.cuh
+++ b/awq_ext/exllamav2/cuda/quant/qdq_5.cuh
--- a/awq_ext/exllamav2/cuda/quant/qdq_6.cuh
+++ b/awq_ext/exllamav2/cuda/quant/qdq_6.cuh
--- a/awq_ext/exllamav2/cuda/quant/qdq_8.cuh
+++ b/awq_ext/exllamav2/cuda/quant/qdq_8.cuh
--- a/awq_ext/exllamav2/cuda/quant/qdq_util.cuh
+++ b/awq_ext/exllamav2/cuda/quant/qdq_util.cuh
--- a/awq_ext/exllamav2/cuda/util.cuh
+++ b/awq_ext/exllamav2/cuda/util.cuh
--- a/awq_ext/exllamav2/ext.cpp
+++ b/awq_ext/exllamav2/ext.cpp
--- a/awq_ext/layernorm/layernorm.cu
+++ b/awq_ext/layernorm/layernorm.cu
--- a/awq_ext/layernorm/layernorm.h
+++ b/awq_ext/layernorm/layernorm.h
--- a/awq_ext/layernorm/reduction.cuh
+++ b/awq_ext/layernorm/reduction.cuh
--- a/awq_ext/position_embedding/pos_encoding.h
+++ b/awq_ext/position_embedding/pos_encoding.h
--- a/awq_ext/position_embedding/pos_encoding_kernels.cu
+++ b/awq_ext/position_embedding/pos_encoding_kernels.cu
--- a/awq_ext/pybind_awq.cpp
+++ b/awq_ext/pybind_awq.cpp
--- a/awq_ext/pybind_awq_ft.cpp
+++ b/awq_ext/pybind_awq_ft.cpp
--- a/awq_ext/pybind_awq_v2.cpp
+++ b/awq_ext/pybind_awq_v2.cpp
--- a/awq_ext/quantization/dequantize.cuh
+++ b/awq_ext/quantization/dequantize.cuh
--- a/awq_ext/quantization/gemm_cuda.h
+++ b/awq_ext/quantization/gemm_cuda.h
--- a/awq_ext/quantization/gemm_cuda_gen.cu
+++ b/awq_ext/quantization/gemm_cuda_gen.cu
--- a/awq_ext/quantization/gemv_cuda.cu
+++ b/awq_ext/quantization/gemv_cuda.cu