© 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. 129588
    Prev
    Next
    INDEX
    Explanations

    access to

    np_max-act · gemini-2.0-flash

    The neuron activates on words that express editing or viewing permissions (e.g. “edit,” “view,” “access,” “allow”) in a user-permissions context.

    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
    bounds
    -0.06
     lines
    -0.06
     Şu
    -0.06
    massage
    -0.06
    Trap
    -0.06
    	score
    -0.06
    viewport
    -0.06
    serie
    -0.06
     میدان
    -0.06
     waived
    -0.06
    POSITIVE LOGITS
     Calendar
    0.08
     (((
    0.07
     kills
    0.07
     Rolling
    0.06
     microscopic
    0.06
    อลล
    0.06
     EM
    0.06
     ((
    0.06
     AHL
    0.06
     QtCore
    0.06
    Activations Density 0.017%

    No Known Activations