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

    Neuronpedia

    INDEX
    Explanations

    conjunctions indicating a contrast or opposing viewpoint

    oai_token-act-pair · gpt-3.5-turbo

    instances of the end-of-text token

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     segment
    -0.70
     nucleus
    -0.60
    built
    -0.58
     ç¥ŀ
    -0.57
     dependent
    -0.55
     incarnation
    -0.54
    ãģ®
    -0.54
    Roaming
    -0.53
     oriented
    -0.53
     Cathedral
    -0.52
    POSITIVE LOGITS
    tons
    1.68
    chers
    1.15
    tery
    1.13
     alas
    1.12
    ter
    1.05
    cher
    1.00
    tern
    0.99
    chery
    0.98
    ts
    0.97
    ler
    0.96
    Act Density 0.102%

    No Known Activations