Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    prepositions indicating a relationship between entities

    oai_token-act-pair · gpt-3.5-turbo

    usage of prepositions related to roles, actions, or legal contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Monsters
    -0.69
    sofar
    -0.66
     Puzzles
    -0.62
    edIn
    -0.61
     Guilty
    -0.60
     apologies
    -0.59
     Oblivion
    -0.56
     Top
    -0.56
     Mund
    -0.56
     Apart
    -0.56
    POSITIVE LOGITS
    -
    1.10
    product
    0.96
    current
    0.95
    usal
    0.93
    growth
    0.92
    sea
    0.91
    pron
    0.91
    etime
    0.90
    orbit
    0.89
    rush
    0.89
    Act Density 0.108%

    No Known Activations