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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating comparison or contrast

    oai_token-act-pair · gpt-3.5-turbo

    phrases including the term "anything" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĥĸ
    -0.73
    sbm
    -0.71
    ãĤ¤ãĥĪ
    -0.68
     Manufacturer
    -0.68
    ãĥ³ãĤ¸
    -0.68
    mud
    -0.65
    aceae
    -0.64
     Gong
    -0.63
     Rumble
    -0.62
    ãĥ£
    -0.60
    POSITIVE LOGITS
     cohesion
    0.64
     leakage
    0.63
     feder
    0.59
     succeeds
    0.59
    assador
    0.59
     happens
    0.59
    ught
    0.58
    Missing
    0.57
    intel
    0.57
     ado
    0.57
    Act Density 0.093%

    No Known Activations