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

    Neuronpedia

    INDEX
    Explanations

    suspicious activities or objects

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "suspicious" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    apsed
    -0.78
    skill
    -0.75
    hung
    -0.75
    mel
    -0.73
    sung
    -0.73
    VEL
    -0.73
    ffen
    -0.72
    produced
    -0.71
    arf
    -0.71
    borg
    -0.70
    POSITIVE LOGITS
    ly
    1.15
     suspicious
    0.99
     suspic
    0.96
    LY
    0.85
    ively
    0.83
    uously
    0.83
     alarms
    0.76
     undermin
    0.75
    lys
    0.74
     detection
    0.74
    Act Density 0.012%

    No Known Activations