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

    Neuronpedia

    INDEX
    Explanations

    percentage values

    oai_token-act-pair · gpt-3.5-turbo

    percentage values represented as numbers followed by the percentage symbol

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     skelet
    -0.83
     shaped
    -0.71
     nurs
    -0.71
     brace
    -0.70
    iden
    -0.69
    pressed
    -0.68
     heav
    -0.68
     compr
    -0.65
     neighb
    -0.65
     appet
    -0.65
    POSITIVE LOGITS
    %.
    1.02
    %%
    0.91
    %,
    0.89
    +.
    0.89
    /-
    0.87
    %-
    0.87
    -+
    0.85
    ooters
    0.83
    %;
    0.83
     ABV
    0.82
    Act Density 0.008%

    No Known Activations