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

    Neuronpedia

    INDEX
    Explanations

    job opportunities and choices

    np_max-act · gemini-2.0-flash

    say "following nouns"

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Negative Logits
     rato
    -0.39
     mano
    -0.36
     larg
    -0.34
    来的
    -0.34
     nove
    -0.34
     largest
    -0.33
     subir
    -0.33
     truc
    -0.33
    edi
    -0.33
    aufbau
    -0.33
    POSITIVE LOGITS
    ful
    0.79
     Seeder
    0.74
     Against
    0.73
    UrlResolution
    0.71
     igång
    0.70
    fulness
    0.68
     Based
    0.68
    extAlignment
    0.67
     Through
    0.66
     Dioxide
    0.66
    Act Density 1.296%

    No Known Activations