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

    Neuronpedia

    INDEX
    Explanations

    following manufacturer instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     घोषित
    0.42
     announced
    0.38
     aggregation
    0.38
     pov
    0.37
    aggregation
    0.37
    reported
    0.37
    衙
    0.37
     aggreg
    0.36
     reported
    0.36
    }|^{
    0.35
    POSITIVE LOGITS
     instructions
    1.99
     Instructions
    1.84
     инструкции
    1.80
    Instructions
    1.78
    instructions
    1.77
     instrucciones
    1.70
     instruction
    1.63
     инструк
    1.63
     instruções
    1.59
     Instruction
    1.48
    Act Density 0.011%

    No Known Activations