Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling.
Stella BidermanHailey SchoelkopfQuentin Gregory AnthonyHerbie BradleyKyle O'BrienEric HallahanMohammad Aflah KhanShivanshu PurohitUSVSN Sai PrashanthEdward RaffAviya SkowronLintang SutawikaOskar van der WalPublished in: ICML (2023)
Keyphrases
- language model
- language modeling
- n gram
- document retrieval
- speech recognition
- language modelling
- retrieval model
- query expansion
- information retrieval
- smoothing methods
- test collection
- probabilistic model
- document ranking
- query terms
- training set
- statistical language models
- context sensitive
- vector space model
- language model for information retrieval
- language models for information retrieval
- ad hoc information retrieval
- retrieval effectiveness
- query specific
- relevance model
- pseudo relevance feedback
- training data
- feature selection