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

    Neuronpedia

    INDEX
    Explanations

    verbs related to completing tasks or operations

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the act of executing tasks or actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    edia
    -0.70
    esa
    -0.66
     magnets
    -0.66
    lua
    -0.66
     Younger
    -0.65
    wi
    -0.64
     Room
    -0.62
    irk
    -0.60
    hetti
    -0.60
    ould
    -0.59
    POSITIVE LOGITS
     reconnaissance
    0.82
    flows
    0.74
     abortions
    0.72
     consultations
    0.71
     surveillance
    0.70
    tan
    0.69
    beit
    0.68
    pez
    0.68
    abad
    0.67
    ghai
    0.66
    Act Density 0.034%

    No Known Activations