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

    Neuronpedia

    INDEX
    Explanations

    marine animals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     underwater
    0.52
     submerged
    0.50
     freshwater
    0.43
     submer
    0.42
     waterline
    0.42
    Underwater
    0.41
     bors
    0.41
     бассе
    0.40
     immersion
    0.39
     swimming
    0.39
    POSITIVE LOGITS
     Clown
    0.83
     clown
    0.81
     puffer
    0.72
     Octopus
    0.70
     anem
    0.66
     Sting
    0.66
     sting
    0.65
     ang
    0.63
     Rays
    0.63
     octopus
    0.62
    Act Density 0.008%

    No Known Activations