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

    Neuronpedia

    INDEX
    Explanations

    questions or prompts in a context

    oai_token-act-pair · gpt-3.5-turbo

    phrases that introduce or transition to a new subject or question

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     donated
    -0.76
     plac
    -0.67
     supported
    -0.63
    PAC
    -0.63
     representations
    -0.63
     preserves
    -0.63
     outweigh
    -0.62
     staffed
    -0.61
     retained
    -0.61
     shelters
    -0.60
    POSITIVE LOGITS
    aceae
    0.79
    ibaba
    0.78
    »Ĵ
    0.76
    culus
    0.76
    ebus
    0.76
    amaz
    0.75
    topic
    0.74
    ultimate
    0.73
     delve
    0.72
    APTER
    0.72
    Act Density 0.475%

    No Known Activations