© 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. 54430
    Prev
    Next
    INDEX
    Explanations

    legal issues

    np_max-act · gemini-2.0-flash

    The neuron detects legal or statutory phrasing that defines prohibited acts or penalties (e.g. “makes it a misdemeanor,” “illegal to,” “unauthorized access”).

    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
    typeorm
    -0.07
     비밀
    -0.07
    numbers
    -0.07
    .addComponent
    -0.07
    refresh
    -0.06
    (BigDecimal
    -0.06
    -т
    -0.06
    Yu
    -0.06
    _prior
    -0.06
    _HINT
    -0.06
    POSITIVE LOGITS
     battled
    0.07
    .Post
    0.06
    adia
    0.06
     дог
    0.06
    估
    0.06
     destin
    0.06
     slur
    0.06
     playlist
    0.06
     hus
    0.06
    件事
    0.06
    Activations Density 0.011%

    No Known Activations