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

    Neuronpedia

    INDEX
    Explanations

    positive evaluations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     imbalances
    0.36
    Obes
    0.34
     blatantly
    0.33
    antiated
    0.32
     черво
    0.31
     разнови
    0.30
    raged
    0.30
     다음과
    0.30
    :',
    0.30
     falsch
    0.30
    POSITIVE LOGITS
     👌
    0.54
     👍
    0.52
    ！
    0.51
    👍
    0.51
    👌
    0.49
     choice
    0.48
    🙌
    0.43
     для
    0.42
     🙌
    0.41
     surv
    0.41
    Act Density 0.061%

    No Known Activations