© 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. 13469
    Prev
    Next
    INDEX
    Explanations

    warnings and safety

    np_max-act · gemini-2.0-flash

    This neuron detects words used in safety and legality warnings—for example “illegal,” “dangerous,” “important,” and similar cautionary 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
    Anime
    -0.07
    _SEQUENCE
    -0.07
    اند
    -0.06
     ACK
    -0.06
    Scalar
    -0.06
    温
    -0.06
    ấn
    -0.06
    047
    -0.06
    Own
    -0.06
     Tu
    -0.06
    POSITIVE LOGITS
     mejorar
    0.07
     sağlam
    0.07
    oolStrip
    0.06
     what
    0.06
    data
    0.06
     ads
    0.06
    _payment
    0.06
     Narendra
    0.06
    makes
    0.06
    /mit
    0.06
    Activations Density 0.078%

    No Known Activations