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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "mostly"

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate the frequency or predominance of an element or concept

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anth
    -0.77
    hawk
    -0.69
    Express
    -0.68
    uers
    -0.67
    Reviewer
    -0.67
    atron
    -0.66
    ciation
    -0.66
    orf
    -0.65
    ilion
    -0.64
    ENA
    -0.64
    POSITIVE LOGITS
     consisted
    0.93
     ceremonial
    0.86
     consisting
    0.84
     unnoticed
    0.83
     unchanged
    0.82
     consist
    0.82
     consists
    0.81
     unexpl
    0.81
     overlooked
    0.79
     focused
    0.78
    Act Density 0.053%

    No Known Activations