Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 129380
    Prev
    Next
    INDEX
    Explanations

    technical language

    np_max-act · gemini-2.0-flash

    The neuron activates on explicit mentions of Facial Action Coding System Action Units (e.g. “AU12,” “lip corner puller,” etc.).

    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
    apt
    -0.07
    acades
    -0.06
    ريق
    -0.06
     squad
    -0.06
     Software
    -0.06
    ,d
    -0.06
    -n
    -0.06
    ني
    -0.06
     Fed
    -0.06
    -ph
    -0.06
    POSITIVE LOGITS
    thresh
    0.07
    <=$
    0.07
    _score
    0.06
    .forEach
    0.06
     İmpar
    0.06
     ↑
    0.06
    .JpaRepository
    0.06
    ();
    ↵
    ↵
    ↵
    0.06
    .RequestMethod
    0.06
    (Query
    0.06
    Activations Density 0.122%

    No Known Activations