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

    Neuronpedia

    INDEX
    Explanations

    referencing or reference

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     คิด
    0.82
     ಪ್ರಯತ್ನ
    0.67
    谥
    0.66
    ุย
    0.63
    謚
    0.63
    ristown
    0.63
    沬
    0.62
    ⎜
    0.62
    諡
    0.61
    anium
    0.61
    POSITIVE LOGITS
     reference
    5.24
     references
    4.99
     Reference
    4.92
    Reference
    4.87
    reference
    4.85
     refer
    4.73
     References
    4.54
    Refer
    4.47
    refer
    4.47
     Refer
    4.46
    Act Density 0.851%

    No Known Activations