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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating something is out of the ordinary or not in line with expectations

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to being out of sync or misaligned with norms or expectations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nai
    -0.84
    incial
    -0.72
    chal
    -0.69
    advertisement
    -0.69
    pei
    -0.68
    ellow
    -0.66
    arnaev
    -0.66
    aman
    -0.65
    ãĤ¶
    -0.63
    è£ħ
    -0.63
    POSITIVE LOGITS
     situations
    0.74
     altogether
    0.70
     anymore
    0.68
     distractions
    0.67
     boredom
    0.63
     syndrome
    0.60
     mouths
    0.60
     entirely
    0.60
    alus
    0.59
     missiles
    0.58
    Act Density 0.085%

    No Known Activations