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

    Neuronpedia

    INDEX
    Explanations

    words related to physical objects

    oai_token-act-pair · gpt-3.5-turbo

    words that denote various types of endings or final touches

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     earthqu
    -0.83
    SIGN
    -0.72
    ISON
    -0.70
    vier
    -0.69
    isons
    -0.65
    VICE
    -0.65
    Effective
    -0.62
    BIL
    -0.61
     Zero
    -0.61
    URRENT
    -0.60
    POSITIVE LOGITS
    hots
    1.29
    omething
    1.09
    pace
    1.05
    hot
    1.05
    tons
    1.04
    creen
    1.03
    peed
    1.02
    cape
    0.98
    mith
    0.96
    uits
    0.95
    Act Density 0.040%

    No Known Activations