Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 130794
    Prev
    Next
    INDEX
    Explanations

    ="@

    np_max-act · gemini-2.0-flash

    The neuron fires on the “@” token that introduces Android XML resource references (in particular those in `android:id="@+id/…"` attributes).

    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
    vat
    -0.08
     wealthiest
    -0.06
    Changing
    -0.06
    відом
    -0.06
     reality
    -0.06
     unravel
    -0.06
     чин
    -0.06
     Reality
    -0.06
     J
    -0.06
    between
    -0.06
    POSITIVE LOGITS
    ="@
    0.08
     シ
    0.08
    ,SLOT
    0.06
     dataSnapshot
    0.06
    ynchronous
    0.06
    اعي
    0.06
    edin
    0.06
     Markup
    0.06
    нивер
    0.06
    fila
    0.06
    Activations Density 0.001%

    No Known Activations