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

    Neuronpedia

    INDEX
    Explanations

    fleeting glimpses and fragments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     solutes
    0.44
     solute
    0.42
     vacu
    0.38
     přímo
    0.37
     variance
    0.36
    了下来
    0.36
     magnifying
    0.36
     sphere
    0.35
    শক্ত
    0.35
     осо
    0.35
    POSITIVE LOGITS
     glimpse
    1.61
     glimpses
    1.61
     glimps
    1.57
     fleeting
    0.94
     झलक
    0.93
     snippets
    0.86
     peek
    0.83
     fragments
    0.81
     hints
    0.81
     fragmentary
    0.81
    Act Density 0.039%

    No Known Activations