© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109652
    Prev
    Next
    INDEX
    Explanations

    silence or minute

    np_max-act · gemini-2.0-flash

    The neuron activates on words naming symbols and rituals of public remembrance, such as “flag,” “moment of silence,” “candles,” “vigil,” and “ceremony.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    保護
    -0.07
    (desc
    -0.06
    恋
    -0.06
    (CON
    -0.06
    Tour
    -0.06
    Rod
    -0.06
     همراه
    -0.06
    訊
    -0.06
    MV
    -0.06
     HID
    -0.06
    POSITIVE LOGITS
     inventions
    0.07
     deactivated
    0.06
     ğ
    0.06
     mentally
    0.06
    phys
    0.06
     Iraq
    0.06
     invokes
    0.06
    产品
    0.06
     Trotsky
    0.06
    :"",↵
    0.06
    Activations Density 0.008%

    No Known Activations