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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating past actions and events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lation
    -0.76
     Translation
    -0.70
    Universal
    -0.67
    ¨
    -0.67
    çīĪ
    -0.67
     endings
    -0.66
    elman
    -0.66
    ankind
    -0.66
    reality
    -0.65
    ~~
    -0.65
    POSITIVE LOGITS
     unlawfully
    1.01
     voluntarily
    0.89
     mol
    0.89
     arrested
    0.86
     illegally
    0.85
     peacefully
    0.84
     felony
    0.83
     abducted
    0.83
     erratic
    0.83
     explosives
    0.82
    Act Density 0.205%

    No Known Activations