pratyushasharma/laser

Issues

• Does LASER have future work?

  #29 opened 7 months ago by pangsg
  2

• After your code is saved, the size of the weights is the same as the pre-trained ones, and no memory is saved. What is the reason for this?

  #28 opened 9 months ago by pursure-D
  6

• How to reproduce Figure 5 analysis in this paper?

  #26 opened 9 months ago by sanyalsunny111
  4

• Reproducing LLAMA-2 metrics

  #27 opened a year ago by sidhantls
  2

• how to get base model accuracy

  #21 opened a year ago by pursure-D
  2

• Problem Encountered During Reproduction

  #22 opened a year ago by ZY123-GOOD
  2

• method of composing reductions across layers

  #19 opened a year ago by KTALS
  2

• Generic model?

  #20 opened a year ago by forresti
  3

• Llama2-7B + TruthfulQA reproduce issue

  #18 opened a year ago by JiwenJ
  8

• Application to three-dimensional tensors

  #17 opened a year ago by xl-lei
  1

• Feature Request for Upcoming Refactoring

  #9 opened a year ago by dkmisra
  0

• what does the 'rate' parameters actually mean in code?

  #14 opened a year ago by KTALS
  2

• Potential improvements for evaluation

  #15 opened a year ago by BenjaminBossan
  1

• Mistral Support

  #4 opened a year ago by fakerybakery
  16

• Rank reduction using random matrix theory

  #12 opened a year ago by fernando-neto-ai
  1

• Do you think it could work for MoE models like Mixtral?

  #7 opened a year ago by BadisG
  2

• Rank-reduced models?

  #8 opened a year ago by turboderp
  4

• Where to Get the Dataset

  #5 opened a year ago by fakerybakery
  5

• Question

  #6 opened a year ago by fakerybakery
  2

• Excellent work, looking forward to following up with further research!

  #1 opened a year ago by caiyuchen-ustc
  3

• License

  #3 opened a year ago by fakerybakery
  6

• What is the ETA on the code

  #2 opened a year ago by MrigankRaman
  2