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

    Neuronpedia

    Jacobian LensNEW
    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. 38647
    Prev
    Next
    INDEX
    Explanations

    Reality TV discussions

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on TV‐show or competition titles and related header tokens (e.g. “MasterChef,” “Total Drama,” “Project Runway,” episode headings, etc.).

    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
    ignored
    -0.07
    Handlers
    -0.06
    Enjoy
    -0.06
     Todo
    -0.06
    逆
    -0.06
    主
    -0.06
    game
    -0.06
    Manchester
    -0.06
    余
    -0.06
     Pang
    -0.06
    POSITIVE LOGITS
     dư�
    0.06
    ेयर
    0.06
     vũ
    0.06
     Wrest
    0.06
     AppleWebKit
    0.06
     Franc
    0.06
    ("%.
    0.06
    -hover
    0.06
     blanket
    0.06
    اكن
    0.06
    Activations Density 0.133%

    No Known Activations