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

    Neuronpedia

    INDEX
    Explanations

    document citations or references

    oai_token-act-pair · gpt-3.5-turbo

    closing brackets or something that denotes the end of segments or lists

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    sung
    -0.79
    ĸļ
    -0.73
    boro
    -0.70
     cradle
    -0.69
    phe
    -0.68
    ãĤ©
    -0.68
     sway
    -0.63
    omore
    -0.63
    userc
    -0.62
    iae
    -0.62
    POSITIVE LOGITS
    ...]
    1.01
    âĢ¦]
    0.84
    TPS
    0.79
    oldemort
    0.75
     Conversely
    0.74
     Flavoring
    0.74
     Alternatively
    0.72
     Uriel
    0.72
    ][
    0.70
    âĨij
    0.70
    Act Density 0.050%

    No Known Activations