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

    Neuronpedia

    INDEX
    Explanations

    terms related to symmetry and patterns

    oai_token-act-pair · gpt-3.5-turbo

    instances of the token "<|endoftext|>" and the sequence "mm"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    GROUND
    -0.84
    dated
    -0.71
    breaks
    -0.71
    grad
    -0.68
    lez
    -0.68
     hazard
    -0.65
    reach
    -0.65
     lawy
    -0.60
    Zip
    -0.60
     hazards
    -0.58
    POSITIVE LOGITS
    useum
    1.08
    mmm
    1.07
    achine
    0.94
    ortal
    0.94
    essage
    0.93
    ittee
    0.93
    andise
    0.89
    etrical
    0.89
    oths
    0.89
    ussen
    0.87
    Act Density 0.024%

    No Known Activations