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

    Neuronpedia

    INDEX
    Explanations

    Refurbished or returned items

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    'habitude
    -0.08
     treaty
    -0.08
     habit
    -0.08
    .
    -0.08
     привыч
    -0.07
    契
    -0.07
    .↵↵
    -0.07
     לאורך
    -0.07
    .]↵↵
    -0.07
     invoking
    -0.07
    POSITIVE LOGITS
     refurbishment
    0.10
     refurb
    0.10
     Sorting
    0.09
     선정
    0.09
     discard
    0.09
    精选
    0.09
     salvage
    0.09
    品質
    0.08
     refurbished
    0.08
     scents
    0.08
    Act Density 0.040%

    No Known Activations