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

    Neuronpedia

    INDEX
    Explanations

    reflexive pronouns and phrases

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating personal responsibility or self-reflection

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     backdrop
    -0.60
    awaru
    -0.58
    milo
    -0.58
    PLA
    -0.57
    Footnote
    -0.55
    Loader
    -0.55
    utenberg
    -0.55
    TN
    -0.53
     Attribution
    -0.53
    pool
    -0.53
    POSITIVE LOGITS
     dearly
    0.76
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    0.75
     wrong
    0.68
     praises
    0.68
     fools
    0.68
    ishers
    0.67
    hest
    0.67
     nightmares
    0.67
    holes
    0.65
     teasp
    0.65
    Act Density 0.257%

    No Known Activations