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

    Neuronpedia

    INDEX
    Explanations

    evaluating content quality and phrasing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    berly
    0.41
    and
    0.41
    but
    0.40
    \|=\
    0.39
    வனாக
    0.39
    light
    0.39
    を用いて
    0.38
    glied
    0.38
    ly
    0.37
    ubern
    0.37
    POSITIVE LOGITS
     этими
    0.54
     هذه
    0.46
     terminology
    0.45
     workflow
    0.44
    局面
    0.42
     aceste
    0.42
     vocab
    0.42
     vocabulary
    0.42
     captcha
    0.41
     conceptos
    0.40
    Act Density 0.036%

    No Known Activations