© 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. 30893
    Prev
    Next
    INDEX
    Explanations

    uncertainty

    np_max-act · gemini-2.0-flash

    This neuron activates on words that signal misrepresentation or a discrepancy between appearance and reality (e.g., false, falsely, truly, represent, confess, do/does).

    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
    iano
    -0.07
     edildi
    -0.07
     elevate
    -0.06
    htm
    -0.06
     Ан
    -0.06
    іблі
    -0.06
     keer
    -0.06
    ποιη
    -0.06
    Authenticate
    -0.06
    mount
    -0.06
    POSITIVE LOGITS
     instincts
    0.07
     Whitney
    0.06
    .Information
    0.06
     logits
    0.06
    Mana
    0.06
    -rock
    0.06
    [::-
    0.06
    อำนวย
    0.06
    ENCE
    0.06
    (ss
    0.06
    Activations Density 0.019%

    No Known Activations