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

    Neuronpedia

    INDEX
    Explanations

    references to problem-solving activities

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "solve" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    treated
    -0.71
    ardy
    -0.68
     lashed
    -0.67
    umption
    -0.67
    oths
    -0.66
    arlane
    -0.65
    acid
    -0.63
     secretaries
    -0.63
    rules
    -0.62
    hee
    -0.62
    POSITIVE LOGITS
     solve
    1.00
    ĵĺ
    0.91
     solving
    0.88
     Puzzles
    0.84
    vable
    0.83
     Mysteries
    0.77
     solves
    0.76
    itaire
    0.76
     puzzles
    0.75
    idian
    0.74
    Act Density 0.019%

    No Known Activations