© 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. 58767
    Prev
    Next
    INDEX
    Explanations

    HTML classes and lists

    np_max-act · gemini-2.0-flash

    This neuron detects HTML/XML markup elements, especially tag names and attributes (e.g. “li,” “class,” “h1,” etc.).

    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
    =%
    -0.07
     противоп
    -0.06
    ap
    -0.06
    .ap
    -0.06
     ErrorMessage
    -0.06
     inade
    -0.06
     pot
    -0.06
    _smooth
    -0.05
    	 ↵
    -0.05
     hints
    -0.05
    POSITIVE LOGITS
     Manager
    0.07
     сайте
    0.07
    iculo
    0.07
    -Length
    0.07
    reservation
    0.07
     vielleicht
    0.07
    (ins
    0.06
     آینده
    0.06
     Sommer
    0.06
    .jp
    0.06
    Activations Density 0.009%

    No Known Activations