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

    Neuronpedia

    INDEX
    Explanations

    is comprised, is telling, is below

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     clientele
    0.88
    各大
    0.85
    地方
    0.80
     circumstances
    0.79
    സൈ
    0.75
     nationalities
    0.75
     consultations
    0.74
     acreage
    0.74
    events
    0.73
     outlets
    0.73
    POSITIVE LOGITS
     incapable
    1.38
     able
    1.36
     unable
    1.34
     banned
    1.30
     unfit
    1.23
     capable
    1.21
     impossible
    1.20
     unsuitable
    1.19
     difficult
    1.18
     devoid
    1.17
    Act Density 0.249%

    No Known Activations