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

    Neuronpedia

    INDEX
    Explanations

    URLs with specific intermediate strings

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    tokens that are parts of web addresses, email/domain names, or filenames (e.g., ".com" and site/filename fragments).

    oai_token-act-pair · gpt-5-miniTriggered by @phamvanluong2003
    New Auto-Interp
    Negative Logits
     ideolog
    -0.83
     notor
    -0.77
     pecuni
    -0.76
     solidar
    -0.73
     indeb
    -0.70
     ristor
    -0.70
     utop
    -0.67
     incess
    -0.66
     lusso
    -0.65
     márm
    -0.64
    POSITIVE LOGITS
    IsContent
    0.73
     shenan
    0.59
     unspeak
    0.58
     exasper
    0.55
     ngl
    0.54
     ineffec
    0.53
     impelled
    0.52
    .*")]
    0.52
     endeav
    0.52
     laboring
    0.51
    Act Density 0.016%

    No Known Activations