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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating additional information or examples

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the term "not to mention."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rend
    -0.84
    sis
    -0.76
    fell
    -0.75
    odes
    -0.75
    olid
    -0.75
    adden
    -0.74
    rete
    -0.74
    rame
    -0.72
    hor
    -0.71
    oward
    -0.71
    POSITIVE LOGITS
     secondly
    0.78
     blah
    0.68
     allergies
    0.68
     condoms
    0.67
     plenty
    0.66
     beware
    0.66
     additionally
    0.65
     importantly
    0.65
     lots
    0.64
     imagine
    0.63
    Act Density 0.100%

    No Known Activations