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

    Neuronpedia

    INDEX
    Explanations

    phrases where a sentiment or statement is reiterated or repeated by different people

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "echo" and its variations, indicating repetition of sentiments or ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    amina
    -0.81
    estate
    -0.79
    atar
    -0.72
    venge
    -0.69
    aho
    -0.67
    ole
    -0.66
    enture
    -0.65
    avez
    -0.64
    uana
    -0.64
    ahime
    -0.64
    POSITIVE LOGITS
     echoes
    1.08
     chorus
    0.88
     echoing
    0.82
     echo
    0.79
     reverber
    0.77
    choes
    0.75
     chamber
    0.74
     chambers
    0.73
    JM
    0.73
     footsteps
    0.73
    Act Density 0.025%

    No Known Activations