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

    Neuronpedia

    INDEX
    Explanations

    referring to specific formats or states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    τρέ
    0.26
    łącz
    0.26
    р
    0.26
     custom
    0.25
    excerpt
    0.25
     compliant
    0.25
     enforce
    0.25
     eme
    0.25
     أنا
    0.24
     intent
    0.24
    POSITIVE LOGITS
    <unused1097>
    0.41
    <unused2040>
    0.41
    <unused642>
    0.40
    <unused1806>
    0.40
     bisschen
    0.39
    <unused674>
    0.39
    䏲
    0.39
    <unused746>
    0.39
    <unused1004>
    0.39
    <unused587>
    0.38
    Act Density 0.000%

    No Known Activations