DylanJHJ's picture
update readme
968c202
metadata
license: apache-2.0

Training detail

  • Training hardware: 2 AMD MI200
  • Batch size: 64
  • Samples per query: 512 (32 * 2 * 8)
  • Data: msmarco-passage 491K
  • Learning steps: 25K with 2.5K warm-up
  • Learning rate: 1e-5