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

    Neuronpedia

    INDEX
    Explanations

    including or listing details

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     worded
    0.55
     hurtful
    0.54
     really
    0.52
     scary
    0.51
     horrible
    0.50
     decirlo
    0.50
     stressful
    0.49
     messed
    0.49
    Honestly
    0.47
     easier
    0.47
    POSITIVE LOGITS
     including
    0.72
     включает
    0.66
     incluindo
    0.63
     частности
    0.63
     зокрема
    0.62
     featuring
    0.61
     Including
    0.61
     включая
    0.61
    including
    0.60
     ,
    0.59
    Act Density 0.000%

    No Known Activations