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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating prison sentences and duration of incarceration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    sentenced to prison

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    EndProject
    -0.39
     silueta
    -0.32
    rungsseite
    -0.32
    zzleHttp
    -0.32
    AppMethodBeat
    -0.31
     Relaciones
    -0.31
     referirse
    -0.31
     ofrecerte
    -0.31
    multicolumn
    -0.30
    Seen
    -0.30
    POSITIVE LOGITS
     jail
    0.92
     prison
    0.88
     jailed
    0.86
     imprisonment
    0.84
     Jail
    0.79
    jail
    0.77
     imprison
    0.72
     imprisoned
    0.71
     Prison
    0.70
     prisión
    0.70
    Act Density 0.029%

    No Known Activations