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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating personal opinions or reflections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cona
    -0.15
    Flush
    -0.15
    ÏĦÏį
    -0.14
    _DOM
    -0.14
    _flush
    -0.14
    lobals
    -0.14
    olla
    -0.13
    Ñijл
    -0.13
    osyal
    -0.13
    NECT
    -0.13
    POSITIVE LOGITS
     think
    0.41
     Think
    0.39
     thinking
    0.39
     THINK
    0.38
    think
    0.38
    Think
    0.36
    thinking
    0.35
     thinks
    0.34
    -thinking
    0.33
     thought
    0.32
    Act Density 0.009%

    No Known Activations