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

    Neuronpedia

    INDEX
    Explanations

    health remedies

    np_max-act · gemini-2.0-flash

    It detects the start of an assistant reply / conversation turn boundary (tokens marking or immediately after the assistant's response start).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     seizure
    -0.08
     dieser
    -0.06
     trainers
    -0.06
    oren
    -0.06
    вод
    -0.06
     мг
    -0.06
     Ik
    -0.06
     bufio
    -0.06
    ingo
    -0.06
    (course
    -0.06
    POSITIVE LOGITS
    esehen
    0.06
    odzi
    0.06
     accessible
    0.06
     coupling
    0.06
     indication
    0.06
     austerity
    0.06
     Portsmouth
    0.06
    asking
    0.06
    .DateFormat
    0.06
    �
    0.06
    Act Density 0.044%

    No Known Activations