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

    Neuronpedia

    INDEX
    Explanations

    questions and documentation

    np_max-act · gemini-2.0-flash

    first-person self-references (I/we) and nearby auxiliary verbs indicating the author’s actions, intentions, or state.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    _bootstrap
    -0.07
    _flight
    -0.07
     Tenant
    -0.07
     gathered
    -0.06
     thống
    -0.06
    bond
    -0.06
     mại
    -0.06
    thalm
    -0.06
    vier
    -0.06
     лим
    -0.06
    POSITIVE LOGITS
     LIABILITY
    0.06
     competing
    0.06
    ValueChanged
    0.06
     industrial
    0.06
     contentValues
    0.06
    /debug
    0.06
     advised
    0.05
    Comparable
    0.05
    GING
    0.05
    '){
↵
    0.05
    Act Density 0.373%

    No Known Activations