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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 57894
    Prev
    Next
    INDEX
    Explanations

    legal proceedings

    np_max-act · gemini-2.0-flash

    The neuron activates on proper names and titled legal terms—e.g. personal names (like “Taft”) and named statutes or roles (like “predator” in “Sexually Violent Predator Act”).

    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
     TAR
    -0.07
    word
    -0.06
    鱼
    -0.06
     stains
    -0.06
    creation
    -0.06
    600
    -0.06
    мо
    -0.06
     maze
    -0.06
    服
    -0.06
     menus
    -0.06
    POSITIVE LOGITS
     podmínek
    0.07
     температура
    0.07
    (groupId
    0.07
     Andrews
    0.06
     resultat
    0.06
     앞
    0.06
     додатков
    0.06
     عدم
    0.06
     TERMS
    0.06
     حالت
    0.06
    Activations Density 0.050%

    No Known Activations