© 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
    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. 6081
    Prev
    Next
    INDEX
    Explanations

    Tell

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of the U.S. military’s “Don’t Ask, Don’t Tell” policy (often shown as DADT or “Don’t Ask, Don’t Tell”).

    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
    딩
    -0.07
     whisk
    -0.06
     setters
    -0.06
    Building
    -0.06
    itors
    -0.06
    atus
    -0.06
    Ind
    -0.06
     alk
    -0.06
    innie
    -0.06
     sanity
    -0.06
    POSITIVE LOGITS
     stereotype
    0.07
     아
    0.07
     Vý
    0.06
     Failed
    0.06
    -known
    0.06
    ERING
    0.06
     prodej
    0.06
     mooie
    0.06
     useful
    0.06
     Artifact
    0.06
    Activations Density 0.007%

    No Known Activations