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

    Neuronpedia

    INDEX
    Explanations

    bulleted list format

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <0xC2>
    0.64
     ­
    0.53
     […]
    0.52
    ­
    0.49
    ̈
    0.49
    […]
    0.47
    ･
    0.46
    ・
    0.45
    <unused61>
    0.43
    <start_of_image>
    0.42
    POSITIVE LOGITS
    <i>
    0.45
    '"
    0.40
    <u>
    0.40
     أنه
    0.39
    albeit
    0.37
     uart
    0.37
     nudge
    0.36
     أل
    0.36
    ,/
    0.35
     fintech
    0.35
    Act Density 0.001%

    No Known Activations