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

    Neuronpedia

    INDEX
    Explanations

    offering other options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     operating
    0.45
     mostly
    0.45
     Each
    0.44
     necessity
    0.43
     (
    0.43
     Operating
    0.43
     strapped
    0.43
     scalability
    0.42
     then
    0.42
     runtime
    0.42
    POSITIVE LOGITS
     інші
    0.55
    říklad
    0.54
     otras
    0.50
    其他的
    0.49
    훨
    0.49
    別の
    0.48
     drugih
    0.48
     ఇతర
    0.48
     ibang
    0.47
     jiné
    0.47
    Act Density 1.323%

    No Known Activations