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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 127160
    Prev
    Next
    INDEX
    Explanations

    Physical places/locations

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on French phrases about opening a bank account, especially references to a bank and its branch (e.g. “chez Banque Populaire” and related banking terms).

    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
     таким
    -0.07
    angle
    -0.07
     calibration
    -0.07
    炉
    -0.06
     B
    -0.06
    plans
    -0.06
     sponsor
    -0.06
    ель
    -0.06
     spends
    -0.06
    abı
    -0.06
    POSITIVE LOGITS
     Stroke
    0.06
    Dealer
    0.06
     ilç
    0.06
    anned
    0.06
    .selectedIndex
    0.06
    �回
    0.06
    ileen
    0.06
    .LinkedList
    0.06
    .Username
    0.06
     injuries
    0.06
    Activations Density 0.020%

    No Known Activations