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

    Neuronpedia

    INDEX
    Explanations

    pre-existing conditions, ethical concerns, gravitational constant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     peanut
    0.49
     regularly
    0.47
     crawled
    0.47
     informant
    0.45
     anthracene
    0.42
     antelope
    0.41
     test
    0.41
     usher
    0.41
     conducting
    0.40
     neural
    0.40
    POSITIVE LOGITS
    uring
    0.48
    Же
    0.47
    assan
    0.46
    possible
    0.46
    atu
    0.45
    ern
    0.45
    atz
    0.45
    ura
    0.44
    Эти
    0.44
    jarige
    0.44
    Act Density 0.002%

    No Known Activations