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

    Neuronpedia

    INDEX
    Explanations

    details relating to identity and transformation

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

    Plot reveals and betrayals

    np_acts-logits-general · gemini-2.0-flash

    reveals or frames actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     meines
    -0.77
     Wicidata
    -0.76
    OGND
    -0.66
    lccn
    -0.63
    ScopeManager
    -0.63
     mojej
    -0.63
     protoimpl
    -0.63
     my
    -0.63
     mijn
    -0.61
    principalColumn
    -0.60
    POSITIVE LOGITS
     reveals
    0.89
     revealing
    0.81
     realizes
    0.80
     confesses
    0.79
     convinces
    0.79
     threatens
    0.78
     flashback
    0.76
     tells
    0.74
     meanwhile
    0.73
     apologizing
    0.72
    Act Density 0.349%

    No Known Activations