© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 126013
    Prev
    Next
    INDEX
    Explanations

    must

    np_max-act · gemini-2.0-flash

    This neuron detects formal requirement statements, specifically the modal “shall” (and related list/numbered formatting) used to specify technical requirements.

    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
    ipay
    -0.07
    anmar
    -0.06
    itors
    -0.06
    Pager
    -0.06
    Lifetime
    -0.06
    ards
    -0.06
    ("("
    -0.06
    期待
    -0.06
     وابسته
    -0.06
    이슈
    -0.06
    POSITIVE LOGITS
     Seven
    0.07
     isConnected
    0.06
     "__
    0.06
    û
    0.06
     justices
    0.06
    (coll
    0.06
    Math
    0.06
     český
    0.06
     Gal
    0.06
     Justice
    0.06
    Activations Density 0.012%

    No Known Activations