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

    Neuronpedia

    INDEX
    Explanations

    cold and hardiness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     metastable
    0.67
    _{
    0.63
     발생
    0.61
     포함
    0.59
    _{\
    0.59
    ।--
    0.58
    ).\
    0.58
    <unused51>
    0.58
     시스템
    0.58
    膛
    0.57
    POSITIVE LOGITS
     differently
    0.61
    ated
    0.53
     herself
    0.51
     Sexuality
    0.49
    jections
    0.49
    ్య
    0.49
     why
    0.49
    atures
    0.49
    ations
    0.48
     visuals
    0.48
    Act Density 0.001%

    No Known Activations