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

    Neuronpedia

    INDEX
    Explanations

    problem-solving

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Jefus
    -0.93
     purpoſe
    -0.72
     myſelf
    -0.71
     ſeveral
    -0.70
     Theſe
    -0.70
     ſtate
    -0.70
     theſe
    -0.68
     himſelf
    -0.67
     themſelves
    -0.67
    ſelves
    -0.66
    POSITIVE LOGITS
     goals
    0.61
    ArgsConstructor
    0.59
     baskets
    0.57
     tanks
    0.55
    snapshots
    0.54
     solvents
    0.52
    shades
    0.52
    istoitu
    0.52
    goals
    0.51
     droplets
    0.50
    Act Density 0.072%

    No Known Activations