Commit History

Auteur SHA1 Bericht Datum
  Nel Nibcord 8f0d19e9b0 embed fix 7 maanden geleden
  Nel Nibcord ee97563b45 Dummied up an abstact save_checkpoint 7 maanden geleden
  Nel Nibcord 02281ebe3d Missed one 7 maanden geleden
  Nel Nibcord dffa17b2d1 Slightly simplified waiting for outstanding requests 7 maanden geleden
  Nel Nibcord 720940d563 Removed statefulModel stuff from mlx impl too 7 maanden geleden
  Nel Nibcord bc2812238f Removed ensure_session to clean stuff up. May revisit later 7 maanden geleden
  Nel Nibcord e9971f74ae Abstract load checkpoint method 7 maanden geleden
  Nel Nibcord 223d35cea0 circular include lol 7 maanden geleden
  Nel Nibcord 8dc73074e6 Nodes don't need an abstract base class 7 maanden geleden
  Nel Nibcord 59af2dd592 Do we need casting here? 7 maanden geleden
  Nel Nibcord b22c21ac16 Some session method cleanup 7 maanden geleden
  Nel Nibcord 98edb393b2 Initialize inference engine session in base class 7 maanden geleden
  Nel Nibcord bcf87e79b7 Okay let's turn no_grad back on. We'll worry about that when tinygrad training works 7 maanden geleden
  Nel Nibcord b7bbda3348 Removed tinygrad StatefulModel class, as it's no longer used 7 maanden geleden
  Nel Nibcord 67f5ae25a5 Fixing tinygrad model 7 maanden geleden
  Nel Nibcord bfa3b36be5 Fixing tinygrad model 7 maanden geleden
  Nel Nibcord 37a75d6b96 Fixing tinygrad model 7 maanden geleden
  Nel Nibcord 0d3abfca95 Made models save properly 8 maanden geleden
  Nel Nibcord 9283f6d7bd Correct loss propagation so we can see the actual loss instead of just the requestor shard's loss 7 maanden geleden
  Nel Nibcord 9eadee310b Basic model saving 8 maanden geleden
  Nel Nibcord 38e368f00b Fixed up the ops so that batches work 8 maanden geleden
  Nel Nibcord dd3d99043b Working distributed training 8 maanden geleden
  Nel Nibcord 175ebc1c42 Coordination biz 8 maanden geleden
  Nel Nibcord 3e869051f6 Okay we should probably await the update 8 maanden geleden
  Nel Nibcord 75c8650f1f Naive network-propagated loss implementation on MLX 7 maanden geleden
  Nel Nibcord 836856824e WIP: Training works on mlx 7 maanden geleden
  Nel Nibcord a6fd7a3430 Generalizing some of the dataset biz while also creating uniform batches 8 maanden geleden
  Nel Nibcord f5efbe1b8f Initial distributed evaluation implementation 7 maanden geleden
  Alex Cheema db9de97fa6 Merge pull request #549 from exo-explore/fixtokenencode 7 maanden geleden
  Alex Cheema c593434808 fix encode endpoint 7 maanden geleden