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

    Neuronpedia

    INDEX
    Explanations

    words related to academic research and scientific studies

    oai_token-act-pair · gemini-2.0-flash

    scientific studies/experiments

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     itſelf
    -0.85
    AndEndTag
    -0.76
    存于互联网档案馆
    -0.76
     ſche
    -0.71
     faſt
    -0.70
     beſt
    -0.68
    󠁿
    -0.67
     houſe
    -0.67
     becauſe
    -0.66
     pleaſure
    -0.65
    POSITIVE LOGITS
    <bos>
    0.73
    colgroup
    0.50
     con
    0.46
     successively
    0.43
     Camp
    0.43
     say
    0.42
     actionMode
    0.42
     fra
    0.41
     couldn
    0.41
    AddField
    0.41
    Act Density 2.782%

    No Known Activations