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

    Neuronpedia

    INDEX
    Explanations

    you might say

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     contain
    0.74
     tell
    0.65
    contain
    0.64
     apply
    0.62
     inform
    0.60
     say
    0.59
     Contains
    0.59
    നിക്ക
    0.59
     notice
    0.58
     grown
    0.58
    POSITIVE LOGITS
     highlighted
    1.54
     indicated
    1.54
     described
    1.52
     asked
    1.47
     depicted
    1.45
     alluded
    1.44
     cited
    1.40
     requested
    1.39
     outlined
    1.38
     said
    1.38
    Act Density 0.188%

    No Known Activations