EricLBuehler/candle-lora

Issues

• Could we have a written walkthrough of finetuning llama/mistral with this?

  #17 opened 8 months ago by Tameflame
  3

• In Llama model, only the embedding layer is converted to lora layer.

  #14 opened 9 months ago by Adamska1008
  5

• Bert model doesn't seem to instantiate with lora weights

  #21 opened 4 months ago by jcrist1
  2

• Installation does not work with newer version of candle (0.6.0)

  #18 opened 4 months ago by czonios
  0

• Can lora also be implemented with stable diffusion?

  #9 opened 4 months ago by staru09
  9

• Is config.rank <= 0 ever true?

  #19 opened 4 months ago by def-roth
  1

• Updated candle-core, candle-nn [0.5.0] release breaks installation of candle-lora and candle-lora-macro dependencies

  #15 opened 8 months ago by Andycharalambous
  0

• Is there any way to save lora-converted model?

  #12 opened 9 months ago by Adamska1008
  5

• How to use canle_lora modle with rust auxm web server

  #10 opened 9 months ago by arthasyou
  4

• error[E0277]: expected a `Fn<(&candle_core::Tensor,)>` closure, found `BatchNorm`

  #8 opened a year ago by EricLBuehler
  0

• Model Merging

  #6 opened a year ago by okpatil4u
  6

• any example for llama_lora training

  #7 opened a year ago by arthasyou
  2

• replace_layer_fields and AutoLoraConvert not working as expected

  #5 opened a year ago by edwin0cheng
  1

• QA-LoRA Implementation and Review

  #3 opened a year ago by okpatil4u
  3

• Add more LoRA transformers

  #4 opened a year ago by EricLBuehler
  0

• Examples for Llama model architecture

  #2 opened a year ago by okpatil4u
  5

• Question: Could we use the same mechanism for Quantization?

  #1 opened a year ago by LLukas22
  4