Model Card for Model ID

llama2-7b-backward-model

Model Description

Finetune the base language model (llama2 7B) with (output, instruction) pairs {(yi, xi)} from the seed data to obtain a backward model Myx := p(x|y). In other words, finetune a model that uses the output to predict the instruction. Use the openassistant-guanaco training set dataset. (25 points)

Framework versions

  • PEFT 0.15.1
Downloads last month
1
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Regiayoung/llama2-7b-backward-model

Adapter
(2342)
this model