© 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. 12779
    Prev
    Next
    INDEX
    Explanations

    past tense placement verbs

    np_max-act · gemini-2.0-flash

    The neuron fires on function-introducing terms in technical/patent prose—words like “provided,” “including,” “having,” or “comprising” that introduce or specify components and their relationships.

    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
    OfWeek
    -0.06
    -double
    -0.06
    amma
    -0.06
    Cors
    -0.06
    embro
    -0.06
     Cors
    -0.06
     Delivery
    -0.06
     {?}
    -0.06
    owl
    -0.06
    (pop
    -0.06
    POSITIVE LOGITS
     parody
    0.07
    MZ
    0.06
     XCTAssert
    0.06
    slu
    0.06
     ballo
    0.06
    431
    0.06
     приход
    0.06
     bracelets
    0.06
     hlavní
    0.06
     byla
    0.06
    Activations Density 0.033%

    No Known Activations