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

    Neuronpedia

    INDEX
    Explanations

    Thought then Action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     repell
    0.41
    ୋ
    0.39
    oppable
    0.38
    ༨
    0.38
    rop
    0.37
    fes
    0.36
     fet
    0.36
    aceae
    0.36
    steel
    0.36
    jal
    0.35
    POSITIVE LOGITS
     Think
    0.46
    nextDouble
    0.42
     Restricted
    0.42
    Think
    0.39
     think
    0.38
     Leverage
    0.38
     Energía
    0.37
     Studying
    0.36
    ff
    0.35
     Leveraging
    0.35
    Act Density 0.009%

    No Known Activations