search

CambioML / pykoi-rlhf-finetuned-transformers

pykoi: Active learning in one unified interface
https://www.cambioml.com
Apache License 2.0
407 stars 43 forks source link

SFT for D2L. DPO #101

Closed llauraa23 closed 8 months ago

llauraa23 commented 9 months ago

Implementation of sft training that also supports d2l application. Implementation of inference evaluation of fine-tuned model. Implementation to support DPO training (on-going).