© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 9017
    Prev
    Next
    INDEX
    Explanations

    deserves/to

    np_max-act · gemini-2.0-flash

    The neuron responds to moral or normative claims—especially phrases expressing that someone “deserves” something (e.g. “deserves to be treated with respect and kindness”).

    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
    dění
    -0.07
    urger
    -0.06
     laboratory
    -0.06
     shortfall
    -0.06
    covers
    -0.06
     демон
    -0.06
    つぶ
    -0.06
    تح
    -0.06
    Acceleration
    -0.06
    緊
    -0.06
    POSITIVE LOGITS
    plevel
    0.07
    aras
    0.06
    iant
    0.06
    _ac
    0.06
    margin
    0.06
     جستارهای
    0.06
    ización
    0.06
    κέ
    0.06
     anim
    0.06
     motiv
    0.06
    Activations Density 0.036%

    No Known Activations