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

    Neuronpedia

    INDEX
    Explanations

    concepts, states, failures, technical terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )`;
    0.49
    olus
    0.47
    )';
    0.46
    )',
    0.46
     Epistle
    0.46
    viridis
    0.46
    )').
    0.45
    *;
    0.45
     inlets
    0.45
     interstices
    0.44
    POSITIVE LOGITS
     Taco
    0.52
     много
    0.50
     ש
    0.50
     Sundance
    0.49
    箓
    0.49
     개
    0.48
    ל
    0.47
    个
    0.47
    И
    0.47
     throws
    0.47
    Act Density 0.001%

    No Known Activations