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

    Neuronpedia

    INDEX
    Explanations

    theft and missing items

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Programming
    -0.08
     comfy
    -0.08
     terlihat
    -0.08
    eleration
    -0.08
     hasa
    -0.08
    enschaft
    -0.08
    .hour
    -0.08
     vrucht
    -0.08
    [...,
    -0.08
     (*.
    -0.08
    POSITIVE LOGITS
     losses
    0.14
     theft
    0.14
     चोरी
    0.13
     pérdidas
    0.13
     loss
    0.12
     pertes
    0.11
    盗
    0.11
     thieves
    0.11
     Theft
    0.11
     Loss
    0.11
    Act Density 0.053%

    No Known Activations