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

    Neuronpedia

    INDEX
    Explanations

    in stages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     depicting
    0.78
     denoting
    0.76
    看待
    0.75
     가진
    0.72
    ပြော
    0.72
     отношениях
    0.71
     portraying
    0.71
    യുള്ള
    0.71
     Among
    0.70
     depictions
    0.70
    POSITIVE LOGITS
     stages
    1.30
    stages
    1.06
     sesuai
    1.03
    Stages
    1.01
     Stages
    0.99
     according
    0.95
     соответствии
    0.95
     accordance
    0.94
     phases
    0.93
    according
    0.91
    Act Density 0.088%

    No Known Activations