© 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. 91974
    Prev
    Next
    INDEX
    Explanations

    ossa

    np_max-act · gemini-2.0-flash

    The neuron activates on self-referential statements about the AI’s own physical or capacity attributes (e.g. “your AI’s physical housing,” “current size,” “capabilities”).

    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
    usch
    -0.06
     clips
    -0.06
     Harmony
    -0.06
    CS
    -0.06
    Test
    -0.06
     Compatibility
    -0.06
    .Line
    -0.06
     restrained
    -0.06
     spins
    -0.06
    肌
    -0.06
    POSITIVE LOGITS
    +n
    0.07
     жел
    0.07
    	JPanel
    0.06
    getView
    0.06
    (graph
    0.06
    Он
    0.06
    .fetchone
    0.06
    *angstrom
    0.06
    $template
    0.06
     Akt
    0.06
    Activations Density 0.001%

    No Known Activations