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

    Neuronpedia

    INDEX
    Explanations

    The neuron is looking for sentences starting with "It is" or "It was" followed by descriptions or declarations

    oai_token-act-pair · gpt-3.5-turbo

    sentence beginnings with "It is" or "It was" followed by statements or declarations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bies
    -0.71
    iling
    -0.71
    mons
    -0.70
     Telescope
    -0.69
    uffs
    -0.65
    apper
    -0.64
     coasts
    -0.63
    aser
    -0.60
     crow
    -0.60
    asers
    -0.59
    POSITIVE LOGITS
     unclear
    0.83
     abundantly
    0.81
     impossible
    0.77
     hardly
    0.75
    yrinth
    0.75
    umo
    0.75
     also
    0.74
    nt
    0.74
     definitely
    0.73
     unlikely
    0.71
    Act Density 0.504%

    No Known Activations