Code

History

Commits on move/encoder_kernel to dev35+
  1. Merge pull request #512 from gordicaleksa/refactor_encoder_bwd_kernelAndrej committed 2 years ago80e0c3b
  2. Remove redundant CPU computationAleksa Gordic committed 2 years ago290c00a
  3. Merge pull request #495 from ChrisDryden/shared_memoryAndrej committed 2 years agoa26041a
  4. Update compile cmd in the dev/cuda READMEAleksa Gordic committed 2 years ago8eafd40
  5. Merge branch 'master' into feature/libdirAndrej Karpathy committed 2 years agof8f9e49
  6. Removed unnecesary shared memory due to blockreduce using static defined shared memoryChristopher committed 2 years ago5450632
  7. Merge pull request #493 from vyom1611/master-2Andrej committed 2 years ago55f3665
  8. Modal benchmarking script updated to replace deprecated callsvyom1611 committed 2 years ago062f096
  9. Merge pull request #476 from lancerts/layernorm-bk-fixAndrej committed 2 years ago2db7716
  10. tweak the logfile parser and visualizerAndrej Karpathy committed 2 years ago0c83a23
  11. amendlancerts committed 2 years ago47c670b
  12. spell explicitly uint to unsigned intlancerts committed 2 years agoc4c985c
  13. fix the issue Mismatch of dweight at layernorm_backward.culancerts committed 2 years agof27ca4d
  14. Minor cleanupRoss Wheeler committed 2 years agobb80f93
  15. Adding dirent.h support to unistd.hRoss Wheeler committed 2 years ago758acdf
  16. bugfixErik Schultheis committed 2 years ago7dc3b7d
  17. fix C % 256 != 0Erik Schultheis committed 2 years ago5f73ecf
  18. fail fast and hard; don't go into the deadlockErik Schultheis committed 2 years agod35daf1
  19. some comments and optimized shared memory amountErik Schultheis committed 2 years ago3b68082
  20. fully vectorized smem accessErik Schultheis committed 2 years ago9eba86f
  21. utilities for Packed and more vectorizationErik Schultheis committed 2 years agocb8cc25
  22. use vectorized access to shared memoryErik Schultheis committed 2 years ago53ee329
  23. fill in the other model sizesAndrej Karpathy committed 2 years agoee6fa5e
  24. simple jupyter notebook example on how to parse and vis the logfilesAndrej Karpathy committed 2 years ago5ec4b77
  25. add ability to export fineweb100B as wellAndrej Karpathy committed 2 years agod0c0143
  26. attempt to fix windows episode 2Andrej Karpathy committed 2 years ago2a0f78d
  27. attempt to fix windows?Andrej Karpathy committed 2 years ago5b96d91
  28. Merge pull request #450 from rosslwheeler/dataloader_win_fixesAndrej committed 2 years ago4ff0412
  29. Merge pull request #439 from lancerts/matmul-fixAndrej committed 2 years ago6a7fd56
  30. fixed mkdir changeRoss Wheeler committed 2 years ago79738d2
  31. Moved windows glob() over to dev/unistd.hRoss Wheeler committed 2 years agoe5083be
  32. Merge branch 'deterministic_layernorm' of https://github.com/ademeure/llm.c into ademeure-deterministic_layernormAndrej Karpathy committed 2 years agodbacaf8
  33. Merge pull request #456 from ngc92/large-batch-fixAndrej committed 2 years agobf03e7f
  34. int -> int64_tErik Schultheis committed 2 years ago1b98637
  35. change default shard size for fineweb10BAndrej Karpathy committed 2 years ago08fe3ee