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

    Neuronpedia

    INDEX
    Explanations

    phrases related to capabilities or abilities

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating capability and potential

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    era
    -0.66
     Bans
    -0.65
    cloth
    -0.64
    udo
    -0.64
    bane
    -0.64
    uren
    -0.62
     Julio
    -0.61
    lights
    -0.60
     Deadline
    -0.60
     Claudia
    -0.60
    POSITIVE LOGITS
     withstand
    0.86
    umbn
    0.75
     conce
    0.75
     discern
    0.75
     inflicting
    0.74
     delivering
    0.74
     navigating
    0.72
     projecting
    0.72
     storing
    0.71
     impart
    0.70
    Act Density 0.072%

    No Known Activations