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

    Neuronpedia

    INDEX
    Explanations

    positive or effective actions and outcomes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    åģ¶
    -0.17
     æ¼Ķ
    -0.16
    heet
    -0.16
    .Utc
    -0.15
    ÙģÙĤ
    -0.15
    uite
    -0.15
    _unused
    -0.14
    uchen
    -0.14
    使
    -0.14
    .ravel
    -0.14
    POSITIVE LOGITS
     reference
    0.28
     mention
    0.26
     notice
    0.23
    reference
    0.21
     note
    0.20
    Reference
    0.19
     witness
    0.18
     referencia
    0.18
     Reference
    0.18
     use
    0.18
    Act Density 0.210%

    No Known Activations