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

    Neuronpedia

    INDEX
    Explanations

    units of measurement and technical specifications

    oai_token-act-pair · gpt-3.5-turbo

    measures and quantities typically found in recipes or technical specifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    istically
    -0.59
     Narc
    -0.57
     Biblical
    -0.56
     lax
    -0.55
     excellence
    -0.55
     murd
    -0.54
     Ninth
    -0.54
     Haram
    -0.54
    ign
    -0.53
     resemb
    -0.53
    POSITIVE LOGITS
    +.
    0.99
    +,
    0.94
    hz
    0.87
    +)
    0.86
    /-
    0.84
    +
    0.81
     apiece
    0.78
    pm
    0.74
    IU
    0.73
    obiles
    0.72
    Act Density 0.056%

    No Known Activations