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

    Neuronpedia

    INDEX
    Explanations

    copy of documents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     copying
    0.70
     copi
    0.60
     imitating
    0.59
     cloning
    0.53
    Clone
    0.52
     brevity
    0.51
     duplicating
    0.50
     copied
    0.50
     clones
    0.50
     imitation
    0.49
    POSITIVE LOGITS
    te
    0.49
    wick
    0.48
    it
    0.47
    。.
    0.46
    se
    0.46
    nable
    0.46
    s
    0.45
    ters
    0.45
    ductory
    0.44
    UIActions
    0.44
    Act Density 0.006%

    No Known Activations