© 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. 12942
    Prev
    Next
    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of event or venue locations—i.e. place names (e.g. halls, hotels, cities) typically following prepositions like “in” or “at.”

    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
     Don
    -0.07
     Caribbean
    -0.07
     Craft
    -0.07
     Tong
    -0.07
    River
    -0.07
    meal
    -0.06
    794
    -0.06
    America
    -0.06
    rip
    -0.06
     retrospect
    -0.06
    POSITIVE LOGITS
    _:*
    0.08
    명을
    0.08
     tabel
    0.07
     lắp
    0.06
    -BEGIN
    0.06
    \Page
    0.06
    -Allow
    0.06
    [MAXN
    0.06
     біля
    0.06
     غر
    0.06
    Activations Density 0.065%

    No Known Activations