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

    Neuronpedia

    INDEX
    Explanations

    terms and phrases indicating defects or shortcomings

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

    detecting flaws and imperfections

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    TemporalType
    -0.69
    Berger
    -0.63
     Infórmanos
    -0.62
    versiti
    -0.60
     Maxi
    -0.58
    Hec
    -0.57
    Maxi
    -0.55
     Nec
    -0.55
     Leger
    -0.54
     RTE
    -0.53
    POSITIVE LOGITS
     flaws
    1.67
     flaw
    1.57
     flawed
    1.48
     flawless
    0.85
     faults
    0.80
     imperfections
    0.77
     imperfection
    0.73
     defectos
    0.69
     weaknesses
    0.69
    faults
    0.68
    Act Density 0.006%

    No Known Activations