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

    Neuronpedia

    INDEX
    Explanations

    objects or items that can be contained or stored in a box

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     suspic
    -0.91
     Haci
    -0.91
     Souha
    -0.88
     Keny
    -0.84
     utaf
    -0.81
     Juf
    -0.80
     philanth
    -0.79
     quoc
    -0.78
     maksi
    -0.78
     sophistic
    -0.78
    POSITIVE LOGITS
     all
    0.96
    all
    0.86
     everything
    0.85
     allemaal
    0.78
     alles
    0.75
    everything
    0.74
     tất
    0.72
     tudo
    0.71
     anything
    0.70
    etc
    0.70
    Act Density 0.424%

    No Known Activations