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

    Neuronpedia

    INDEX
    Explanations

    potential for ambiguity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ourselves
    0.41
     reconstructed
    0.40
     our
    0.40
     matters
    0.40
     midnight
    0.40
     Florida
    0.40
     fake
    0.39
     titles
    0.39
     оригіналу
    0.38
     दिवंगत
    0.38
    POSITIVE LOGITS
    셩
    0.43
     skillset
    0.43
    Stdin
    0.42
    Cent
    0.40
    并将
    0.40
    agility
    0.39
    Skills
    0.39
    Cooldown
    0.39
    cam
    0.38
    Unicode
    0.38
    Act Density 0.002%

    No Known Activations