© 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. 68559
    Prev
    Next
    INDEX
    Explanations

    direction/orientation

    np_max-act · gemini-2.0-flash

    The neuron activates on phrases that specify the direction of an action—e.g. “in the direction of the grain/finish/scales”—marking instructions about how to wipe, brush, or rub along a surface.

    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
    люд
    -0.08
    ندق
    -0.07
    欢
    -0.07
     allocator
    -0.06
     �
    -0.06
    ucc
    -0.06
    ОН
    -0.06
    (mapping
    -0.06
    coe
    -0.06
    Handled
    -0.06
    POSITIVE LOGITS
    CRM
    0.07
    .Nome
    0.07
    ]="
    0.06
    ेट
    0.06
    VRT
    0.06
     retail
    0.06
    Delivery
    0.06
     prostituerade
    0.06
    ="$(
    0.06
    ]{
    0.06
    Activations Density 0.121%

    No Known Activations