© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 8768
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that mark C++ generic/template type usages—e.g. type parameters (T, Int), scope separators (::), and angle-bracket syntax.

    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
    Page
    -0.06
    atem
    -0.06
     Pages
    -0.06
    \DependencyInjection
    -0.06
    _characters
    -0.06
    .epoch
    -0.06
    Approval
    -0.06
     MyApp
    -0.06
     अज
    -0.06
     Komm
    -0.06
    POSITIVE LOGITS
    _DEPTH
    0.07
    мага
    0.06
    不
    0.06
     إليه
    0.06
    灭
    0.06
    üph
    0.06
    ehler
    0.06
     conferences
    0.06
    lage
    0.06
    (ir
    0.06
    Activations Density 0.074%

    No Known Activations