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

    Neuronpedia

    INDEX
    Explanations

    qualitative descriptors

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     prone
    0.46
     estremamente
    0.46
     extremamente
    0.43
     அதிகம்
    0.42
     Much
    0.41
     Significantly
    0.40
    Worth
    0.40
     Remarkably
    0.40
     真
    0.40
     Enough
    0.39
    POSITIVE LOGITS
     concerted
    0.47
    簡単な
    0.41
     expanded
    0.40
     nefarious
    0.38
     heightened
    0.38
     evolved
    0.38
     cursory
    0.37
    具体的な
    0.37
     sizable
    0.37
     fruitless
    0.37
    Act Density 0.671%

    No Known Activations