path # lines of code include/SIMD.h 470 include/BinAlgo.h 64 include/Algo-Direct2.h 215 include/AlgoXCodes.h 23 include/Portable.h 157 include/Algo-Direct-Common.h 281 include/AAlloc.h 63 include/Type.h 171 include/BinSearch.h 7 scripts/stale.py 33 benchmarking/switchback/info_a100_py2.jsonl 60 benchmarking/switchback/make_plot_with_jsonl.py 126 benchmarking/switchback/speed_benchmark.py 134 csrc/cpu_ops.h 7 csrc/common.h 18 csrc/common.cpp 18 csrc/cpu_ops.cpp 48 csrc/pythonInterface.cpp 314 csrc/kernels.cuh 99 csrc/ops.cuh 157 csrc/kernels.cu 3505 csrc/mps_ops.h 1 csrc/ops.cu 740 csrc/mps_kernels.metal 103 csrc/mps_ops.mm 54 pyproject.toml 54 setup.py 36 bitsandbytes/autograd/_functions.py 420 bitsandbytes/autograd/__init__.py 1 bitsandbytes/cextension.py 67 bitsandbytes/__main__.py 3 bitsandbytes/optim/adagrad.py 116 bitsandbytes/optim/__init__.py 16 bitsandbytes/optim/lion.py 162 bitsandbytes/optim/adam.py 322 bitsandbytes/optim/rmsprop.py 98 bitsandbytes/optim/optimizer.py 532 bitsandbytes/optim/lars.py 169 bitsandbytes/optim/sgd.py 89 bitsandbytes/optim/lamb.py 95 bitsandbytes/optim/adamw.py 178 bitsandbytes/__init__.py 17 bitsandbytes/functional.py 2202 bitsandbytes/diagnostics/__init__.py 1 bitsandbytes/diagnostics/main.py 64 bitsandbytes/diagnostics/utils.py 7 bitsandbytes/diagnostics/cuda.py 109 bitsandbytes/research/autograd/_functions.py 274 bitsandbytes/research/autograd/__init__.py 1 bitsandbytes/research/__init__.py 6 bitsandbytes/research/nn/__init__.py 1 bitsandbytes/research/nn/modules.py 63 bitsandbytes/cuda_specs.py 27 bitsandbytes/utils.py 126 bitsandbytes/nn/__init__.py 18 bitsandbytes/nn/modules.py 541 bitsandbytes/nn/triton_based_modules.py 178 bitsandbytes/triton/dequantize_rowwise.py 53 bitsandbytes/triton/quantize_rowwise.py 55 bitsandbytes/triton/__init__.py 1 bitsandbytes/triton/int8_matmul_rowwise_dequantize.py 171 bitsandbytes/triton/triton_utils.py 3 bitsandbytes/triton/quantize_global.py 105 bitsandbytes/triton/int8_matmul_mixed_dequantize.py 171 bitsandbytes/triton/quantize_columnwise_and_transpose.py 63 bitsandbytes/consts.py 10 install_cuda.py 83 check_bnb_install.py 14 _typos.toml 8