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

    Neuronpedia

    INDEX
    Explanations

    phrases related to casual conversations and interactions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of casual conversation and humor

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    AMD
    -0.53
     agric
    -0.52
    GPU
    -0.51
    ordes
    -0.51
    quartered
    -0.48
    orsi
    -0.47
     lapt
    -0.47
     reliant
    -0.47
    Firstly
    -0.47
    products
    -0.47
    POSITIVE LOGITS
     fuckin
    0.87
     fucking
    0.73
     uh
    0.68
     gonna
    0.67
     bitch
    0.65
    eeee
    0.64
     fucked
    0.63
     wanna
    0.61
     kinda
    0.60
     gotta
    0.58
    Act Density 1.404%

    No Known Activations