Commit History

Author SHA1 Message Date
  Eric J. Wang af30df1999 Unified tokenizer update PR (#146) 3 years ago
  Martin Thissen d3760cd84a Added fine-tuned7b model for German language (#134) 3 years ago
  Thaweewat 6853b8802e Add Thai weight URL on READ.ME (#132) 3 years ago
  Eric Wang fcdb143f1f Amend README 3 years ago
  Eric Wang 72aabcb5a4 Remove LLaMA download code, as a precaution 3 years ago
  Eric Wang 8955a9c5a1 bos, eos in generate.py 3 years ago
  Eric J. Wang 1384a4d24c Update README.md for multi-GPU training 3 years ago
  bofeng huang c7eabb86e2 Add french version "vigogne" (#127) 3 years ago
  Eric J. Wang a74793c571 Rearrange resources on README, add 13B-30B models 3 years ago
  Eric Wang b12c3b90f8 Unwind input masking to avoid confusion 3 years ago
  Eric Wang e04897baae fix fp16 inference 3 years ago
  Eric J. Wang 052da42cbb Replace Colab with HF in README 3 years ago
  Eric Wang 7fb06c6c22 Revert "Mask out prompt tokens for real" 3 years ago
  Eric Wang 2204a71505 set EPOCHS back to 3 3 years ago
  Eric Wang 4a712d4d8e Mask out prompt tokens for real 3 years ago
  Eric Wang fac53721a2 masking bugfix 3 years ago
  Eric J. Wang 3cdbfe5b0c Update README.md 3 years ago
  Eric J. Wang c08c34eabb mention chatbot project in README.md 3 years ago
  Eric J. Wang f0082d8e8b Link to resources more prominently 3 years ago
  Eric J. Wang d38802e843 Point volunteers to Open Assistant 3 years ago
  Kohaku-Blueleaf b5a1a0bca7 Add support for valid set size 0 (#83) 3 years ago
  Kohaku-Blueleaf 0af44f0262 Add option for output dir (#84) 3 years ago
  Kohaku-Blueleaf 450206caaf Fix torch.compile call on windows (#81) 3 years ago
  Karun 81eb72f707 cleans up alphabetical prompts (#76) 3 years ago
  Eric Wang 997f6cd81f slider for tokens generated 3 years ago
  Eric Wang cfad895aa1 mask prompt in loss 3 years ago
  Eric J. Wang d66908c0ca Remove messy test code 3 years ago
  Yaqub Mahmoud 0e752ea5f3 Update requirements.txt (#67) 3 years ago
  Eric Wang c83e30ab78 generate.py tweaks 3 years ago
  Eric Wang 80fd9833db don't share publicly 3 years ago