Code

History

Commits on feature/managed235+
  1. i misspelled reducedAndrej Karpathy committed 2 years ago18298f3
  2. reduce across GPUs nicerAndrej Karpathy committed 2 years ago8c586f9
  3. fallback to memory allocation of m,v,master_weights on host automatically in case of OOM. will run slower but won't OOMAndrej Karpathy committed 2 years agoe6856bc
  4. fix makefile for multigpu setupsAndrej Karpathy committed 2 years ago2882ec6
  5. hint to hostErik Schultheis committed 2 years agof72c1f2
  6. just try to allocate on device; fallback if that failsErik Schultheis committed 2 years agoc845757
  7. fall back to cudaMallocManaged for optimizer states if we're out of memoryErik Schultheis committed 2 years ago0d52d2a
  8. Merge pull request #740 from karpathy/gordicaleksa-fix_dataloader2Andrej committed 2 years ago4c84bc7
  9. fix tokenizer omgAndrej Karpathy committed 2 years ago755458d
  10. attempt to fix PRAndrej Karpathy committed 2 years ago16635d4
  11. Merge branch 'fix_dataloader2' of https://github.com/gordicaleksa/llm.c into gordicaleksa-fix_dataloader2Andrej Karpathy committed 2 years ago9740a65
  12. Merge pull request #738 from ademeure/faster_compileAndrej committed 2 years ago1787210
  13. Refactor tokenize, delegate insteadAleksa Gordic committed 2 years ago68624cc
  14. Refactor magic/version header dictAleksa Gordic committed 2 years agoe8739c8
  15. Improve compile time via nvidia-smi instead of __nvcc_device_query (& add "--threads=0").ademeure committed 2 years agofc88344
  16. Improve compile time via nvidia-smi instead of __nvcc_device_query (& add "--threads=0").ademeure committed 2 years ago9b1c14b
  17. Replace fire with argparseAleksa Gordic committed 2 years agoeb1b359
  18. Add fire to reqsAleksa Gordic committed 2 years ago311c149
  19. Add fineweb llama 3 supportAleksa Gordic committed 2 years ago5076563
  20. Use fire module, update readmeAleksa Gordic committed 2 years agof8c1ff5
  21. Add tiny shakespeareAleksa Gordic committed 2 years ago15b76bf
  22. Change default dataset path to tiny storiesAleksa Gordic committed 2 years agob018b0c
  23. Update header descriptionAleksa Gordic committed 2 years ago71419cb
  24. Add llama 3 tokenization and data loaderAleksa Gordic committed 2 years agod3151c4
  25. Merge pull request #725 from gordicaleksa/llamaAndrej committed 2 years ago6e6a528
  26. Remove unnecessary commentAleksa Gordic committed 2 years agod773c88
  27. Add comments, fix stop tokensAleksa Gordic committed 2 years agoc5c87fc
  28. Add data loader not implemented exceptionAleksa Gordic committed 2 years ago8b672ff
  29. Address Andrej's PR commentsAleksa Gordic committed 2 years agof1c91f8
  30. Remove pycacheAleksa Gordic committed 2 years ago89addd3
  31. Remove the whole llmc_pyAleksa Gordic committed 2 years agoac01536
  32. Remove prompts.jsonAleksa Gordic committed 2 years agodfd459b
  33. Rename the new file to train llama3Aleksa Gordic committed 2 years ago624ed3c
  34. Add llama config error checkAleksa Gordic committed 2 years agob7c98c9
  35. Add explicit external maskAleksa Gordic committed 2 years agob25e325