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

    Neuronpedia

    INDEX
    Explanations

    words related to evidence or documentation

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ild
    -0.60
    las
    -0.58
    éĸ
    -0.55
    chwitz
    -0.55
    mun
    -0.54
    othe
    -0.53
    ãĤ§
    -0.53
    ror
    -0.53
    ron
    -0.52
    irable
    -0.52
    POSITIVE LOGITS
    theless
    0.59
     everywhere
    0.58
     consisted
    0.54
     Everywhere
    0.50
     onwards
    0.49
     lasted
    0.47
     consists
    0.45
     notwithstanding
    0.44
     extensively
    0.44
     misled
    0.43
    Act Density 1.294%

    No Known Activations