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

    Neuronpedia

    INDEX
    Explanations

    error

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _decay
    -0.07
     Damn
    -0.07
     pops
    -0.07
    fasta
    -0.07
    finance
    -0.07
     diplomats
    -0.06
     popped
    -0.06
     neighbourhood
    -0.06
     consultant
    -0.06
    。但
    -0.06
    POSITIVE LOGITS
    ->__
    0.07
     Typed
    0.07
    .Items
    0.07
     hend
    0.07
    hes
    0.06
     errors
    0.06
    _RS
    0.06
    =$((
    0.06
    PER
    0.06
     TT
    0.06
    Act Density 0.023%

    No Known Activations