© 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. 7928
    Prev
    Next
    INDEX
    Explanations

    Food

    np_max-act · gemini-2.0-flash

    This neuron activates on coarse insults or profanities.

    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
    -find
    -0.07
     họp
    -0.06
    _sm
    -0.06
     moeten
    -0.06
     irm
    -0.06
     pharmacist
    -0.06
    .cart
    -0.06
     elderly
    -0.06
    awl
    -0.06
     subtraction
    -0.06
    POSITIVE LOGITS
    centage
    0.06
    κα
    0.06
     penetrate
    0.06
     çift
    0.06
    테
    0.06
    sock
    0.06
     يا
    0.06
    mmas
    0.06
     konuda
    0.06
    ouple
    0.06
    Activations Density 0.005%

    No Known Activations