© 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. 44193
    Prev
    Next
    INDEX
    Explanations

    px

    np_max-act · gemini-2.0-flash

    The neuron is primarily firing on HTML/CSS code tokens—especially tag names and class/ID selectors you see in stylesheet or markup (e.g. “div”, “.content”, “.resume-wrapper”, 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
    _not
    -0.06
    needle
    -0.06
    ğ
    -0.06
    Inject
    -0.06
     dotenv
    -0.06
     neon
    -0.06
     employer
    -0.06
     посл
    -0.06
    接受
    -0.06
    715
    -0.06
    POSITIVE LOGITS
     duro
    0.07
    革
    0.06
     Rouge
    0.06
    agues
    0.06
     Variant
    0.06
    .GetSize
    0.06
    ("<?
    0.06
     considered
    0.06
    _PROTOCOL
    0.06
    _thread
    0.06
    Activations Density 0.023%

    No Known Activations