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

    Neuronpedia

    INDEX
    Explanations

    phrases related to explosions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "explode" or its variations such as "exploding" or "exploded"

    oai_token-act-pair · gpt-4-turbo

    terms and variations related to explosive actions or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    the word "explode" or its variations

    oai_token-act-pair · gpt-4-turbo
    New Auto-Interp
    Negative Logits
    ña
    -0.77
    abet
    -0.75
    esthetic
    -0.73
    leg
    -0.72
    aunder
    -0.72
    uth
    -0.70
    chal
    -0.70
    ajo
    -0.65
    ittee
    -0.64
    trained
    -0.64
    POSITIVE LOGITS
     explodes
    0.95
     bursting
    0.88
     exploding
    0.84
     exploded
    0.80
     explode
    0.79
     explosion
    0.76
    opl
    0.74
     burst
    0.73
     shells
    0.73
    alore
    0.73
    Act Density 0.011%

    No Known Activations