© 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. 109629
    Prev
    Next
    INDEX
    Explanations

    his

    np_max-act · gemini-2.0-flash

    This neuron activates on third-person possessive pronouns (e.g., “his,” “her,” “’s”) indicating personal or biographical references.

    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
    .setBackgroundColor
    -0.07
    \Table
    -0.07
    _reward
    -0.07
    ้นท
    -0.06
    UTC
    -0.06
     reminders
    -0.06
    ยน
    -0.06
    _side
    -0.06
    _fsm
    -0.06
     sizeof
    -0.06
    POSITIVE LOGITS
    自己的
    0.06
     레벨
    0.06
     Lum
    0.06
     окруж
    0.06
     '~/
    0.06
     tractor
    0.06
     ImageButton
    0.06
     waiter
    0.06
     incest
    0.06
     rig
    0.06
    Activations Density 0.027%

    No Known Activations