Commits · 490153b29fc86dcf9c074f64faa7af92d0a33f73 · OpenDAS / bitsandbytes

10 Jul, 2023 9 commits
- Added generation tests. · 490153b2
  Tim Dettmers authored Jul 10, 2023
  
  490153b2
- Added ARCH guard for bfloat16 computations. · 1c774ece
  Tim Dettmers authored Jul 10, 2023
  
  1c774ece
- Fixed typo in cuda_install.sh. · 0a1cced3
  Tim Dettmers authored Jul 10, 2023
  
  0a1cced3
- Changelog and version bump. · 0d344b70
  Tim Dettmers authored Jul 10, 2023
  
  0d344b70
- Fixed Makefile and added CUDA 12.2 install. · 73aa4e0a
  Tim Dettmers authored Jul 10, 2023
  
  73aa4e0a
- Merge remote-tracking branch 'origin/inference' · 5f492d43
  Tim Dettmers authored Jul 10, 2023
  
  5f492d43
- Merge pull request #469 from shadeMe/linear-layer-device · 196d6f5d
  Tim Dettmers authored Jul 10, 2023
```
Add `device` parameter to `Linear` subclasses and `Embedding`
```
  196d6f5d
- Added fp32 compute type for gemv_4bit. · 5fab6734
  Tim Dettmers authored Jul 09, 2023
  
  5fab6734
- Added test for Param4bit.to() and fixed double quant behavior. · cef519c8
  Tim Dettmers authored Jul 09, 2023
  
  cef519c8
09 Jul, 2023 4 commits
- Fixed a bug where gemv_4bit would return a wrongly sized tensor. · 6a905be5
  Tim Dettmers authored Jul 09, 2023
  
  6a905be5
- Added double quantization support and tests. · 0f0390ac
  Tim Dettmers authored Jul 09, 2023
  
  0f0390ac
- Added FP4 fast inference support. · 94168d79
  Tim Dettmers authored Jul 09, 2023
  
  94168d79
- Added abitrary data types; fixed a bug for small matrices. · 4b88d69d
  Tim Dettmers authored Jul 09, 2023
  
  4b88d69d
08 Jul, 2023 2 commits
- Turning optimization (float accumulation). 185 vs 50. · eefbf602
  Tim Dettmers authored Jul 08, 2023
  
  eefbf602
- Added warp_shuffle indexing 185 vs 54. · 7e49b5b9
  Tim Dettmers authored Jul 08, 2023
  
  7e49b5b9
05 Jul, 2023 1 commit
- Added bfloat16 quantizations and tests. · 02fd80cb
  Tim Dettmers authored Jul 04, 2023
  
  02fd80cb
04 Jul, 2023 2 commits
- Vectorized loads, conflict free NF4; 52 vs 172. · dfe6900b
  Tim Dettmers authored Jul 04, 2023
  
  dfe6900b
- Initial 4-bit naive batch size 1, 81 vs 185. · f89ff93e
  Tim Dettmers authored Jul 03, 2023
  
  f89ff93e
20 Jun, 2023 1 commit
- Release 0.39.1. · 4395d68c
  Tim Dettmers authored Jun 19, 2023
  
  4395d68c
19 Jun, 2023 1 commit
- Merge pull request #503 from TimDettmers/efficient_8bit_serialize · 2d321a75
  Tim Dettmers authored Jun 19, 2023
```
Make 8-bit serialization more memory-efficient (v2)
```
  2d321a75
14 Jun, 2023 1 commit
- Only rearrange weight if it exists · b599fdb1
  Max Ryabinin authored Jun 14, 2023
  
  b599fdb1
09 Jun, 2023 3 commits
- Rearrange the weights directly in state dict before loading · c1f3f56d
  Max Ryabinin authored Jun 09, 2023
  
  c1f3f56d
- Improve memory efficiency of 8-bit serialization · f734076e
  Max Ryabinin authored Jun 09, 2023
  
  f734076e
- Extract get_tile_inds to a separate function · 4fb37d45
  Max Ryabinin authored Jun 09, 2023
  
  4fb37d45
01 Jun, 2023 2 commits
- Add `device` parameter to `Embedding` · db49ad43
  shadeMe authored Jun 01, 2023
  
  db49ad43
- Add `device` parameter to `Linear` subclasses · 9cac5dd1
  shadeMe authored Jun 01, 2023
  
  9cac5dd1
31 May, 2023 3 commits
- Added debugging functions. · e54d2730
  Tim Dettmers authored May 30, 2023
  
  e54d2730
- Added lookup table. · b7f04e2a
  Tim Dettmers authored May 30, 2023
  
  b7f04e2a
- Added changes for deployment. · ac5550a0
  Tim Dettmers authored May 30, 2023
  
  ac5550a0
24 May, 2023 3 commits
- Bumped version. · 0f40fa3f
  Tim Dettmers authored May 23, 2023
  
  0f40fa3f
- Added PagedLion and bf16 Lion. · 1b8772a8
  Tim Dettmers authored May 23, 2023
  
  1b8772a8
- Fixed Makefile. · 2bce175d
  Tim Dettmers authored May 23, 2023
  
  2bce175d
07 May, 2023 4 commits
- Fixed gradient accumulation test. · 4bd11518
  Tim Dettmers authored May 07, 2023
  
  4bd11518
- Merge remote-tracking branch 'origin/main' into merge · 675baa79
  Tim Dettmers authored May 07, 2023
  
  675baa79
- Fixed prefetch bug for non-paged tensors; added benchmark. · f64cfe65
  Tim Dettmers authored May 06, 2023
  
  f64cfe65
- Changed prefetching. · 41a9c708
  Tim Dettmers authored May 06, 2023
  
  41a9c708
06 May, 2023 2 commits
- Added paged optimizers. · 44d68ff2
  Tim Dettmers authored May 06, 2023
  
  44d68ff2
- Added paging. · ec38ba95
  Tim Dettmers authored May 06, 2023
  
  ec38ba95
02 May, 2023 2 commits
- 4-bit draft; 128 vector load 240. · 264a9485
  Tim Dettmers authored May 02, 2023
  
  264a9485
- Warp multi-specialization 240. · 869b7e83
  Tim Dettmers authored May 02, 2023
  
  869b7e83