Commit History

Autor SHA1 Mensaxe Data
  Matthias Reso 8f5db330de Correct model urls for llama3 in tests hai 10 meses
  Matthias Reso ab254d121c 7b -> 8b hai 10 meses
  Matthias Reso cad284c66f Replace new model url hai 10 meses
  Matthias Reso 8b0a233c1a Use new chat format in custom dataset hai 10 meses
  Matthias Reso fac41298b0 Adapt test_custom_dataset to new model hai 10 meses
  Matthias Reso 960014a3bb Fix test_custom_dataset by introducing a stable sort algorithm hai 10 meses
  Matthias Reso 147aaa29bc Remove deprecated pytest_cmdline_preparse hai 1 ano
  Matthias Reso 0022d97163 remove decapoda-research/llama-7b-hf tokenizer and skip tests if meta-llama/Llama-2-7b is not available hai 1 ano
  Matthias Reso 53fd82355f Add missing changes hai 1 ano
  Matthias Reso 8620ab8ac2 Fix invalid labels for context in custom dataset/oasst1 hai 1 ano
  Matthias Reso 52c417b7d5 Merge branch 'fix/invalidate_label_for_chat' into feature/length_based_batch_sampling hai 1 ano
  Matthias Reso eafea7b366 Invalidate labels in dialog dataset to disable loss hai 1 ano
  Matthias Reso 10f9367e56 fix missing labels in datasets hai 1 ano
  Matthias Reso ca41c1c697 Adjust tests to len based batch sampling hai 1 ano
  Matthias Reso eccf7cb3dc Check additional example in resulting dataset hai 1 ano
  Matthias Reso ec00a2f722 Fix batching error hai 1 ano
  Matthias Reso dc507b4e55 Finish implementation oasst preprocessing for of custom dataset hai 1 ano
  Matthias Reso 26b9b7dbb2 Give an explicit error message if custom datset function is not found hai 1 ano
  Matthias Reso 9cb3fddfcf Move datasets tests in their own subfolder hai 1 ano