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

    Neuronpedia

    INDEX
    Explanations

    phrases related to actions or events being done or occurring

    oai_token-act-pair · gpt-3.5-turbo

    past participle verbs and their variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    adra
    -0.94
    adan
    -0.80
    matically
    -0.70
    efully
    -0.70
    ively
    -0.66
    ptive
    -0.65
    hang
    -0.63
    pered
    -0.63
    strength
    -0.63
    doi
    -0.61
    POSITIVE LOGITS
    nesday
    1.03
    escription
    0.97
     Ones
    0.89
    irect
    0.88
    dit
    0.85
    paren
    0.82
    irection
    0.82
    monton
    0.82
    own
    0.81
    ragon
    0.80
    Act Density 0.145%

    No Known Activations