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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 54860
    Prev
    Next
    INDEX
    Explanations

    FIG

    np_max-act · gemini-2.0-flash

    This neuron detects references to figures or numbered items (e.g., “FIG. 1,” “FIGS. 1–2,” and other numeric labels).

    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
    guarded
    -0.08
    _CARD
    -0.07
    _db
    -0.07
    нять
    -0.07
    _RCC
    -0.06
    Hat
    -0.06
     Hurt
    -0.06
    ulnerable
    -0.06
     beled
    -0.06
     Emperor
    -0.06
    POSITIVE LOGITS
    िव
    0.07
     likely
    0.07
    ":"","
    0.06
     singapore
    0.06
    旅行
    0.06
    erli
    0.06
     coquine
    0.06
    ею
    0.06
    .comment
    0.06
    {//
    0.06
    Activations Density 0.009%

    No Known Activations