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

    Neuronpedia

    INDEX
    Explanations

    temperature scaling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     accumulates
    0.46
    ל
    0.45
     ansatz
    0.45
     lanterns
    0.43
     injective
    0.42
     abiotic
    0.42
    י
    0.42
     analogues
    0.42
     glimpses
    0.42
     accumulations
    0.41
    POSITIVE LOGITS
     Americans
    0.52
     Let
    0.51
    ны
    0.50
     Canada
    0.50
     Florida
    0.49
    ভাষায়
    0.48
     Department
    0.48
     Miss
    0.47
     新
    0.47
    fees
    0.46
    Act Density 0.004%

    No Known Activations