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. 1542
    Prev
    Next
    INDEX
    Explanations

    articles

    np_max-act · gemini-2.0-flash

    This neuron fires on singular noun phrases introduced by the indefinite articles “a” or “an.”

    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
     Castillo
    -0.07
    -corner
    -0.07
    Third
    -0.07
    血
    -0.07
     kapı
    -0.06
     мі
    -0.06
    اجر
    -0.06
     matched
    -0.06
    ieber
    -0.06
     BEFORE
    -0.06
    POSITIVE LOGITS
     نگ
    0.06
     Adobe
    0.06
     primitives
    0.06
    ="#">
    0.06
     Authenticate
    0.06
     detal
    0.06
     ύ
    0.06
     inserting
    0.05
     narratives
    0.05
     serialized
    0.05
    Activations Density 0.067%

    No Known Activations