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

    Neuronpedia

    INDEX
    Explanations

    script, maintain, promql

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.46
    髗
    0.45
     смогли
    0.44
    :::
    0.41
    >∕</
    0.38
    葲
    0.38
    TestSuite
    0.37
    ంథ
    0.37
    滯
    0.37
    尔多
    0.36
    POSITIVE LOGITS
     playback
    0.43
     paced
    0.42
     rate
    0.42
     rates
    0.39
     Zim
    0.39
     played
    0.37
     paid
    0.36
    playable
    0.36
     traced
    0.36
     hires
    0.36
    Act Density 0.001%

    No Known Activations