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

    Neuronpedia

    INDEX
    Explanations

    provides final answer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     caldera
    0.37
     سسٹم
    0.37
    റ്
    0.37
    Rough
    0.36
     Forma
    0.36
     coy
    0.35
    koop
    0.35
     perist
    0.35
     basaltes
    0.35
     …,
    0.35
    POSITIVE LOGITS
    anyakan
    0.38
    4
    0.34
    assertArg
    0.34
     다양
    0.33
    <0xBE>
    0.33
    च्या
    0.33
    ensuremath
    0.33
    mathbb
    0.33
    antu
    0.33
    agot
    0.33
    Act Density 0.005%

    No Known Activations