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

    Neuronpedia

    INDEX
    Explanations

    long sentences

    np_max-act · gemini-2.0-flash

    Tokens marking conversation structure and speaker/header metadata (e.g., header IDs, role labels like user/assistant, and named-speaker tokens).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
     comple
    -0.06
    ient
    -0.06
    according
    -0.06
     tantal
    -0.06
     skeptical
    -0.06
     animate
    -0.06
    oky
    -0.06
     SVM
    -0.06
    Expect
    -0.06
    かり
    -0.06
    POSITIVE LOGITS
     بق
    0.07
     Σχ
    0.07
     "")
↵
    0.07
    .Middle
    0.07
    :date
    0.07
    .SetFloat
    0.07
     Blocked
    0.06
    .KeyCode
    0.06
     ")
↵
    0.06
     çarp
    0.06
    Act Density 0.059%

    No Known Activations