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

    Neuronpedia

    INDEX
    Explanations

    bonus, bonuses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     focuses
    0.42
     ഉണ്ടാ
    0.42
     broadcasters
    0.41
     alerted
    0.40
     understand
    0.39
     differentiated
    0.39
     fills
    0.38
    washed
    0.38
     broadcaster
    0.37
     understandably
    0.37
    POSITIVE LOGITS
     τε
    0.43
     ПП
    0.42
     prolog
    0.41
     δι
    0.40
    並
    0.39
    琹
    0.38
    碐
    0.38
    0.37
    дри
    0.37
     लैपटॉप
    0.36
    Act Density 0.000%

    No Known Activations