Commits · 84fb7e9808171dbf6b4790fd4044c2fd39aee09c · OpenDAS / AutoAWQ

07 Sep, 2023 1 commit
- Support Falcon 180B · 475481b4
  Casper authored Sep 07, 2023
  
  475481b4
06 Sep, 2023 4 commits
- Remove simple dispatch · 562c0d52
  Casper authored Sep 06, 2023
  
  562c0d52
- Multi-GPU support for quantized models · f8273a0c
  Casper Hansen authored Sep 06, 2023
  
  f8273a0c
- Get correct devices · 7cf0d987
  Casper Hansen authored Sep 06, 2023
  
  7cf0d987
- Correct input sizes · ab7d68e7
  Casper Hansen authored Sep 06, 2023
  
  ab7d68e7
05 Sep, 2023 4 commits
- Remove CustomQuantAttention class · 42655843
  Casper Hansen authored Sep 05, 2023
  
  42655843
- Revert "Update neox kernel" · bbe1d46a
  Casper Hansen authored Sep 05, 2023
```
This reverts commit 4fe9974a.
```
  bbe1d46a
- Update neox kernel · 4fe9974a
  Casper Hansen authored Sep 05, 2023
  
  4fe9974a
- Attempt new module · 8bb8fe20
  Casper Hansen authored Sep 05, 2023
  
  8bb8fe20
04 Sep, 2023 2 commits
- Pass trust_remote_code to AutoModelForCausalLM · 0090ad81
  Casper Hansen authored Sep 04, 2023
  
  0090ad81
- fuse_layers bug fix · 9106ce29
  qwopqwop200 authored Sep 04, 2023
  
  9106ce29
02 Sep, 2023 5 commits
- Fix model references · 0aa4a596
  Casper Hansen authored Sep 02, 2023
  
  0aa4a596
- Refactor MPT Quant MLP · 1df0136e
  Casper Hansen authored Sep 02, 2023
  
  1df0136e
- Refactor Llama Quant MLP · ded3ea71
  Casper Hansen authored Sep 02, 2023
  
  ded3ea71
- Refactor Llama Quant RMSNorm · 620966e8
  Casper Hansen authored Sep 02, 2023
  
  620966e8
- Refactor Llama quant attention · 2082197d
  Casper Hansen authored Sep 02, 2023
  
  2082197d
27 Aug, 2023 1 commit
- Rename to hijacked_inputs · 50d1025f
  Casper authored Aug 27, 2023
  
  50d1025f
25 Aug, 2023 3 commits
- fixed catcher input name · 6b1c96c7
  EC2 Default User authored Aug 25, 2023
  
  6b1c96c7
- updated max_new_tokens_key · fac1af55
  EC2 Default User authored Aug 25, 2023
  
  fac1af55
- added gptj · 86fcf708
  EC2 Default User authored Aug 25, 2023
  
  86fcf708
24 Aug, 2023 1 commit
- Integrate fused modules into AWQ model loading · 63346c34
  Casper Hansen authored Aug 24, 2023
  
  63346c34
22 Aug, 2023 2 commits
- Implement loading correct sequence length based on config + custom max_new_tokens · d9bab50c
  Casper Hansen authored Aug 22, 2023
  
  d9bab50c
- Set full quant as default. Always load unquantized with AutoModel instead of accelerate. · 321d74ff
  Casper Hansen authored Aug 22, 2023
  
  321d74ff
19 Aug, 2023 10 commits
- Support Bloom models · 5c2c85fe
  Casper Hansen authored Aug 19, 2023
  
  5c2c85fe
- Fix logic for saving model · dd3010fb
  Casper Hansen authored Aug 19, 2023
  
  dd3010fb
- Support safetensors · 04e177ad
  Casper Hansen authored Aug 19, 2023
  
  04e177ad
- Support Falcon 7B+40B · d73d13b2
  Casper Hansen authored Aug 19, 2023
  
  d73d13b2
- Fix dict access · 06073073
  Casper Hansen authored Aug 19, 2023
  
  06073073
- Save model file with filename from quant_config · dcd58818
  Casper Hansen authored Aug 19, 2023
  
  dcd58818
- Create generate method · cc779bed
  Casper Hansen authored Aug 19, 2023
  
  cc779bed
- Load quantized model with saved quant_config · d68441d0
  Casper Hansen authored Aug 19, 2023
  
  d68441d0
- Make BaseAWQForCausalLM a torch.nn.Module · 84d23089
  Casper Hansen authored Aug 19, 2023
  
  84d23089
- Rename q_config -> quant_config. Include w_bit in quant_config. Save quant_config.json. · d35ade75
  Casper Hansen authored Aug 19, 2023
  
  d35ade75
18 Aug, 2023 6 commits
- Add OPT support · 6f30f051
  Casper Hansen authored Aug 18, 2023
  
  6f30f051
- Increase robustness of model loading · 471f811b
  Casper Hansen authored Aug 18, 2023
  
  471f811b
- Fix comment · 916fdf97
  Casper Hansen authored Aug 18, 2023
  
  916fdf97
- Refactor model class. Add LLaMa support (LLaMa-2, Vicuna, etc). · e2858cef
  Casper Hansen authored Aug 18, 2023
  
  e2858cef
- save_quantized working in all cases, from_quantized adapted. · d430694b
  Casper Hansen authored Aug 18, 2023
  
  d430694b
- Implemented save_quantized. Generalize from_quantized.Add comments. · ed347704
  Casper Hansen authored Aug 18, 2023
  
  ed347704
17 Aug, 2023 1 commit
- Implement from_pretrained. Fix static methods and classmethods · 14d198c6
  Casper Hansen authored Aug 17, 2023
  
  14d198c6