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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 84804
    Prev
    Next
    INDEX
    Explanations

    business correspondence

    np_max-act · gemini-2.0-flash

    This neuron identifies structural/chat metadata tokens—e.g. speaker‐ID markers, header delimiters, and NAME_x placeholders.

    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
    сол
    -0.07
    ом
    -0.06
     zoning
    -0.06
     akıl
    -0.06
    -grand
    -0.06
    [last
    -0.06
     LGBTQ
    -0.06
    hra
    -0.06
    리가
    -0.06
    List
    -0.06
    POSITIVE LOGITS
     استفاده
    0.06
     peaceful
    0.06
    Exclude
    0.06
    最後
    0.06
     smoothed
    0.06
    ンティ
    0.06
    :↵↵↵
    0.06
    (Spring
    0.06
     poids
    0.06
     необходим
    0.06
    Activations Density 0.054%

    No Known Activations