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

    Neuronpedia

    INDEX
    Explanations

    hidden patterns relationships

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     compliant
    0.56
     intuitively
    0.55
     defensively
    0.54
     agnostic
    0.50
     smiled
    0.46
     slightly
    0.46
     truncate
    0.46
     aligned
    0.45
     decimated
    0.44
     eloquently
    0.44
    POSITIVE LOGITS
     परिस्थिती
    0.69
     أو
    0.67
    ácter
    0.67
    ischem
    0.67
     relationships
    0.66
     beliefs
    0.65
     ҡ
    0.65
    ală
    0.65
    zogen
    0.64
    ksjon
    0.64
    Act Density 1.107%

    No Known Activations