© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126979
    Prev
    Next
    INDEX
    Explanations

    code headers and copyright

    np_max-act · gemini-2.0-flash

    The neuron detects lines of comment‐block metadata—e.g. the leading “/*” license or copyright header lines in source code.

    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
     HID
    -0.07
     unordered
    -0.06
    _Internal
    -0.06
    ＝
    -0.06
    .Linear
    -0.06
     satire
    -0.06
    ném
    -0.06
          
    -0.06
    .dispatch
    -0.06
    НО
    -0.06
    POSITIVE LOGITS
    раст
    0.08
    cold
    0.07
     mapa
    0.07
    -inf
    0.07
     hoop
    0.06
    POINTS
    0.06
     grandes
    0.06
     بگ
    0.06
     된다
    0.06
     Ein
    0.06
    Activations Density 0.003%

    No Known Activations