© 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. 110470
    Prev
    Next
    INDEX
    Explanations

    Autobiographical descriptions

    np_max-act · gemini-2.0-flash

    This neuron activates on character-profile or background details—especially phrases that introduce personal origins and demographic info (e.g. “from a middle-class background,” ages, places).

    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
    .PNG
    -0.06
    んでいる
    -0.06
    .Angle
    -0.06
    622
    -0.06
    ?!
    -0.06
     GREAT
    -0.06
    LOCATION
    -0.06
    Medium
    -0.06
     sincerity
    -0.06
     METHOD
    -0.06
    POSITIVE LOGITS
    locking
    0.07
    iker
    0.07
    orpor
    0.07
    (chalk
    0.06
    abar
    0.06
    yro
    0.06
    cter
    0.06
    -order
    0.06
    _math
    0.06
    rom
    0.06
    Activations Density 0.010%

    No Known Activations