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. 130696
    Prev
    Next
    INDEX
    Explanations

    NSA surveillance

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of U.S. intelligence‐and‐surveillance entities or programs (e.g. NSA, surveillance programs, national security agency).

    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
                                                                   
    -0.07
                                                          
    -0.06
    ляем
    -0.06
     طریق
    -0.06
    >{@
    -0.06
     mar
    -0.06
     STATE
    -0.06
    _X
    -0.06
    Checker
    -0.06
    Pa
    -0.06
    POSITIVE LOGITS
    Void
    0.06
    ssl
    0.06
     largely
    0.06
     sağlan
    0.06
    /use
    0.06
     abilities
    0.06
    Recovered
    0.06
    Texture
    0.06
    uste
    0.06
    ео
    0.06
    Activations Density 0.006%

    No Known Activations