© 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. 8760
    Prev
    Next
    INDEX
    Explanations

    -router

    np_max-act · gemini-2.0-flash

    The neuron activates on routing-related terms (e.g. “router,” “react‐router,” “Route”), so it’s detecting mentions of the app’s routing library or route components.

    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
    ۱۰
    -0.06
    Profiles
    -0.06
    ご
    -0.06
    _threads
    -0.06
     Ke
    -0.06
    137
    -0.06
    do
    -0.06
     arguably
    -0.06
    LogLevel
    -0.06
    ILON
    -0.06
    POSITIVE LOGITS
    -router
    0.09
    _FB
    0.08
     ин
    0.07
    identified
    0.07
     Router
    0.07
    Router
    0.07
     obsessed
    0.07
     longstanding
    0.07
     staggering
    0.06
     SOFTWARE
    0.06
    Activations Density 0.001%

    No Known Activations