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

    Neuronpedia

    INDEX
    Explanations

    studies and respondents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    你应该
    0.50
     conceivable
    0.49
     useful
    0.48
     computationally
    0.46
     syntactic
    0.46
     propagation
    0.45
     성능
    0.45
     excitation
    0.45
     zeroth
    0.45
     unavoid
    0.45
    POSITIVE LOGITS
     respondents
    1.03
     surveyed
    1.02
     Respondents
    1.00
     survey
    0.99
     Survey
    0.98
     encuesta
    0.98
     सर्वेक्षण
    0.95
     Surveys
    0.95
    survey
    0.93
    Survey
    0.92
    Act Density 0.015%

    No Known Activations