Commits · 1e8ea70b267575f3392a96b195f39af3ddbddbe2 · mirrored_repos / MachineLearning / meta-llama / Llama Recipes

This project is mirrored from https://github.com/meta-llama/llama-recipes. Pull mirroring failed 1 hour ago.
Last successful update 3 months ago.

Aug 25, 2023
- adding llama code inference · 1e8ea70b
  Hamid Shojanazeri authored 1 year ago
  
  1e8ea70b
Aug 17, 2023
- bugfix: remove duplicate load_peft_model · fcc817e9
  hongbo.mo authored 1 year ago
  
  fcc817e9
Aug 08, 2023
- moving Bt to the try block · 51269b81
  Hamid Shojanazeri authored 1 year ago
  
  51269b81
Aug 07, 2023
- clean up and typo fixes · 754b5d22
  Hamid Shojanazeri authored 1 year ago
  
  754b5d22
Aug 04, 2023
- Add BT after peft module load · bf4e0a13
  Hamid Shojanazeri authored 1 year ago
  
  bf4e0a13
- adding flash attention and xformer memory efficient through PT SDPA · 44ef280d
  Hamid Shojanazeri authored 1 year ago
  
  44ef280d
Aug 02, 2023
- add max_pad length as an arg · 9b0eae40
  Hamid Shojanazeri authored 1 year ago
  
  9b0eae40
Aug 01, 2023
- adding padding to inference script with embedding size changing · 6beadcd8
  Hamid Shojanazeri authored 1 year ago
  
  6beadcd8
Jul 31, 2023
- clean up · c4e96af6
  Hamid Shojanazeri authored 1 year ago
  
  c4e96af6
Jul 25, 2023
- remove the unused code · ceaf6301
  Hamid Shojanazeri authored 1 year ago
  
  ceaf6301
Jul 22, 2023
- adding inference updates · 7ad5bd5e
  Hamid Shojanazeri authored 1 year ago
  
  7ad5bd5e
Jul 20, 2023
- aliging special tokens in toeknizer with HF latest · 7ec390bf
  Hamid Shojanazeri authored 1 year ago
  
  7ec390bf
- adding dist checkpoint handler for signle gpu · 15053268
  Hamid Shojanazeri authored 1 year ago
  
  15053268
Jul 19, 2023
- adding fsdp checkpoint loading · 25b60ee8
  Hamid Shojanazeri authored 1 year ago
  
  25b60ee8
- adding time and tp config and new pad token · 37dac0eb
  Hamid Shojanazeri authored 1 year ago
  
  37dac0eb
Jul 17, 2023
- Initial commit · 4767f09e
  chauhang authored 1 year ago
  
  4767f09e