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

    Neuronpedia

    INDEX
    Explanations

    verbs in past tense

    oai_token-act-pair · gpt-3.5-turbo

    forms of the verb "to be" and related expressions of existence or condition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    acca
    -0.87
    kay
    -0.79
    eah
    -0.74
    ampa
    -0.71
    GW
    -0.70
    rongh
    -0.69
    onga
    -0.65
    orses
    -0.63
    OB
    -0.62
    arez
    -0.62
    POSITIVE LOGITS
     nonetheless
    1.39
     nevertheless
    1.21
     retaining
    0.87
     retained
    0.83
     retains
    0.80
     rarely
    0.79
     still
    0.77
     cautioned
    0.75
     seldom
    0.75
     alas
    0.72
    Act Density 0.163%

    No Known Activations