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

    Neuronpedia

    INDEX
    Explanations

    action verbs related to offering assistance or support

    oai_token-act-pair · gpt-3.5-turbo

    terms related to assistance and improvement in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     forbids
    -0.64
     distur
    -0.62
     presided
    -0.61
    ****************
    -0.60
    âĨ
    -0.58
    Flor
    -0.57
     interrupts
    -0.56
    never
    -0.56
     consisting
    -0.55
     denotes
    -0.55
    POSITIVE LOGITS
     safely
    0.84
    escape
    0.80
     efficiently
    0.76
     healthier
    0.75
     navigate
    0.72
     sustainable
    0.70
     better
    0.70
     smarter
    0.69
     manageable
    0.68
    uate
    0.68
    Act Density 0.269%

    No Known Activations