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

    Neuronpedia

    INDEX
    Explanations

    references to the publication of research studies and papers

    oai_token-act-pair · gpt-3.5-turbo

    instances of publishing dates and journal references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    zik
    -0.78
     awoken
    -0.73
    gage
    -0.72
    resa
    -0.69
    anka
    -0.69
    outer
    -0.69
    rious
    -0.68
    rendered
    -0.68
    rouse
    -0.67
     caller
    -0.67
    POSITIVE LOGITS
    DragonMagazine
    0.93
     Bei
    0.78
    Else
    0.77
     Proceedings
    0.75
    çīĪ
    0.72
     Courier
    0.71
     redacted
    0.70
     Paste
    0.70
    Journal
    0.70
    NAS
    0.67
    Act Density 0.124%

    No Known Activations