© 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 EvalsExportsinterp-engineNEWAPI 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. 57579
    Prev
    Next
    INDEX
    Explanations

    expressing opinions or knowledge

    np_max-act · gemini-2.0-flash

    This neuron responds to first‐person opinion statements—particularly tokens like “I,” “it’s,” and similar self-referential constructions indicating the speaker’s personal view.

    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
    AB
    -0.07
    _ARG
    -0.06
     "${
    -0.06
    -simple
    -0.06
     poco
    -0.06
    (elements
    -0.06
    xFA
    -0.06
     vicious
    -0.06
     Ferrari
    -0.06
    �
    -0.06
    POSITIVE LOGITS
    dispatch
    0.06
    βέρ
    0.06
     scope
    0.06
    libc
    0.06
     Mason
    0.06
     ahead
    0.06
    мож
    0.06
    lops
    0.06
     realms
    0.06
     проп
    0.06
    Activations Density 0.023%

    No Known Activations