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

    Neuronpedia

    INDEX
    Explanations

    introduces clauses or explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     solved
    0.47
    0
    0.45
     movies
    0.44
     pancakes
    0.43
     lasagna
    0.43
    ю
    0.42
    ations
    0.42
     varied
    0.41
     infinito
    0.41
    हारिक
    0.41
    POSITIVE LOGITS
     Whereas
    0.60
     Firstly
    0.60
     Within
    0.57
     Referring
    0.56
     Whenever
    0.55
     Namely
    0.55
     While
    0.54
     When
    0.54
     Upon
    0.53
     In
    0.52
    Act Density 0.005%

    No Known Activations