r-three

AttriBoT

AttriBoT: A Bag of Tricks for Efficiently Approximating Leave-One-Out Context Attribution

common-pile

Code for collecting, processing, and preparing datasets for the Common Pile

fib

git-theta

git extension for {collaborative, communal, continual} model development

mats

phatgoose

Code for PHATGOOSE introduced in "Learning to Route Among Specialized Experts for Zero-Shot Generalization"

RAD

Reference implementation for Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model

realistic_evaluation_of_model_merging_for_compositional_generalization

smear

t-few

Code for T-Few from "Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning"

r-three/t-few

Code for T-Few from "Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning"

r-three/common-pile

Code for collecting, processing, and preparing datasets for the Common Pile

r-three/git-theta

git extension for {collaborative, communal, continual} model development

r-three/phatgoose

Code for PHATGOOSE introduced in "Learning to Route Among Specialized Experts for Zero-Shot Generalization"

r-three/RAD

Reference implementation for Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model

r-three/mats

r-three/smear

r-three/fib

r-three/realistic_evaluation_of_model_merging_for_compositional_generalization

r-three/AttriBoT

AttriBoT: A Bag of Tricks for Efficiently Approximating Leave-One-Out Context Attribution

r-three/MetaICL

An original implementation of "MetaICL Learning to Learn In Context" by Sewon Min, Mike Lewis, Luke Zettlemoyer and Hannaneh Hajishirzi

r-three/llama

Inference code for LLaMA models

r-three/MixEval

The official evaluation suite and dynamic data release for MixEval.