Commits · 8645d1f71cc78155887bc3ba082b1a610a05e31f · OpenDAS / bitsandbytes

30 Mar, 2023 1 commit
- Added normal quant. · 8645d1f7
  Tim Dettmers authored Mar 29, 2023
  
  8645d1f7
27 Mar, 2023 1 commit
- Some small changes. · 69810521
  Tim Dettmers authored Mar 27, 2023
  
  69810521
05 Feb, 2023 3 commits
- Fixed matmul_fp4 transpose. · c361f842
  Tim Dettmers authored Feb 05, 2023
  
  c361f842
- Added matmul_fp4 to the benchmark. · cfe4705e
  Tim Dettmers authored Feb 04, 2023
  
  cfe4705e
- Forward matmul_fp4 tests pass. · 160a8358
  Tim Dettmers authored Feb 04, 2023
  
  160a8358
04 Feb, 2023 1 commit
- Added fp4 quant/dequant and dequant optimizations. · 3ac5840c
  Tim Dettmers authored Feb 04, 2023
  
  3ac5840c
19 Nov, 2022 1 commit
- Fixed k-bit quantization maps. · eb028e6e
  Tim Dettmers authored Nov 19, 2022
  
  eb028e6e
08 Nov, 2022 1 commit
- Fixed bug in cpu quant; faster GPU dequant. · 08fa2e7b
  Tim Dettmers authored Nov 07, 2022
  
  08fa2e7b
07 Nov, 2022 2 commits
- Fixed blockwise test and logic. · e0e697b1
  Tim Dettmers authored Nov 06, 2022
  
  e0e697b1
- Added blocksizes 2048, 1024, and 512 to blockwise quant. · 6bc2b992
  Tim Dettmers authored Nov 06, 2022
  
  6bc2b992
06 Nov, 2022 3 commits
- Added k<256 quantile estimate. · 2f2063ba
  Tim Dettmers authored Nov 06, 2022
  
  2f2063ba
- Added k-bit fp8 map. · 98cbc4bc
  Tim Dettmers authored Nov 06, 2022
  
  98cbc4bc
- Added k-bit linear quantization. · caf18325
  Tim Dettmers authored Nov 06, 2022
  
  caf18325
04 Nov, 2022 1 commit
- Added FP8 quantization map. · 1efb87d8
  Tim Dettmers authored Nov 03, 2022
  
  1efb87d8
27 Oct, 2022 1 commit

Simplify statements into equivalent, modern variants · 0b078403

Tom Aarsen authored Oct 27, 2022

via pyupgrade --py37-plus. The changes e.g. are subclassing from object, calling super() with super(ThisClass, self), or old-style syntax formatting.

0b078403

24 Oct, 2022 1 commit
- Isolated CUDASetup logging; all tests green. · df86625a
  Tim Dettmers authored Oct 24, 2022
  
  df86625a
13 Sep, 2022 1 commit
- Fixed cpu blockwise quantization for small input tensors. · c05dd42d
  Tim Dettmers authored Sep 13, 2022
  
  c05dd42d
11 Sep, 2022 1 commit
- Fixed 2^31 max size issue for cpu blockwise quant. · 19a7adca
  Tim Dettmers authored Sep 11, 2022
  
  19a7adca
23 Aug, 2022 1 commit
- Some initial code. Needs to be tested. · 7e0fb655
  Tim Dettmers authored Aug 23, 2022
  
  7e0fb655
16 Aug, 2022 2 commits
- Added fused bias in dequant_mm. · dede3430
  Tim Dettmers authored Aug 16, 2022
  
  dede3430
- Removed storage() from get_ptr; added boilerplate for bias dequant_mm. · 1ed2fa2f
  Tim Dettmers authored Aug 16, 2022
  
  1ed2fa2f
01 Aug, 2022 2 commits
- reran black with linelength 80 for greater readability · ea7c14f8
  Titus von Koeller authored Aug 01, 2022
  
  ea7c14f8
- ran black and isort for coherent code formatting · bfa0e332
  Titus von Koeller authored Aug 01, 2022
  
  bfa0e332
27 Jul, 2022 2 commits
- Added col_ampere outlier extraction kernel. · 32fa459e
  Tim Dettmers authored Jul 26, 2022
  
  32fa459e
- Working outlier extraction for Turing. · bcab99ec
  Tim Dettmers authored Jul 26, 2022
  
  bcab99ec
26 Jul, 2022 2 commits
- Boilerplate and test for extract_outliers. · cbb901ac
  Tim Dettmers authored Jul 26, 2022
  
  cbb901ac
- Removed rowscale (segfaults on ampere). · 1e88edd8
  Tim Dettmers authored Jul 25, 2022
  
  1e88edd8
25 Jul, 2022 1 commit
- Fixed makefile; fixed Ampere igemmlt_8 bug. · 8b1fd32e
  Tim Dettmers authored Jul 25, 2022
  
  8b1fd32e
22 Jul, 2022 1 commit
- Most tests passing. · c771b3a7
  Tim Dettmers authored Jul 22, 2022
  
  c771b3a7
06 Oct, 2021 1 commit
- Initial commit · 74399248
  Tim Dettmers authored Oct 05, 2021
  
  74399248