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

    Neuronpedia

    INDEX
    Explanations

    texts denoted by Ċ that introduce a new section or paragraph

    oai_token-act-pair · gpt-3.5-turbo

    structured discussions about various topics or issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     disapprove
    -0.67
    boro
    -0.67
     equip
    -0.64
     manoeuv
    -0.64
     withdraw
    -0.63
    undle
    -0.62
    utical
    -0.62
    utic
    -0.61
     inactive
    -0.61
     Vald
    -0.61
    POSITIVE LOGITS
    ³³³
    1.07
    ³³³³
    1.06
    ³³³³³³³³³³³³³³³³
    0.99
    ³³³³³³³³
    0.91
    ³³
    0.88
    Consider
    0.84
    Firstly
    0.81
    Reason
    0.81
    Anyway
    0.81
    Specifically
    0.80
    Act Density 0.675%

    No Known Activations