Skip to content

v0.2.2

Compare
Choose a tag to compare
@yifanmai yifanmai released this 31 Mar 18:20
00c8eab

Models

  • Added Cohere Command (#1321)
  • Added Flan-T5 (#1398)
  • Added H3 (#1398)
  • Added GPT-NeoXT-Chat-Base-20B (#1407)
  • Added OpenAI gpt-3.5-turbo-0301 (#1401)
  • Added AI21 Jurassic-2 models (#1409)

Scenarios

  • Some improvements to LEXTREME and LexGLUE legal scenarios (#1429)
  • Added OpinionsQA scenario (#1424)

Metrics

  • Added multilabel classification metrics (#1408)

Framework

  • Fixed --exit-on-error not working and added --skip-completed-runs (#1400)
  • Disabled tqdm in non-interactive mode (#1351)
  • Added plotting (#1403, #1411)
  • Added Hugging Face Model Hub integration (#1103)

Evaluations

  • Added evaluation results for Cohere Command and Aleph Alpha Luminous