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

    Neuronpedia

    INDEX
    Explanations

    flawed, biased, corrupted, fake

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    l
    0.55
    langan
    0.53
    en
    0.49
    o
    0.47
     Clouds
    0.46
    Computed
    0.46
    el
    0.45
     ermöglichen
    0.45
    lan
    0.44
    et
    0.44
    POSITIVE LOGITS
     inanimate
    0.45
     tiver
    0.45
    inizin
    0.44
     healed
    0.44
     undead
    0.44
     widowed
    0.43
     receptive
    0.43
     poking
    0.42
    '-
    0.42
     layak
    0.42
    Act Density 0.029%

    No Known Activations