© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 14102
    Prev
    Next
    INDEX
    Explanations

    one

    np_max-act · gemini-2.0-flash

    The neuron activates on the word “one” (and its synonym “single”) when used to emphasize a single unit or location (e.g. “one place,” “one hub,” “one interface,” “single day”).

    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
    ersistence
    -0.08
     presumption
    -0.08
    Access
    -0.07
     teaches
    -0.06
    Suite
    -0.06
     quello
    -0.06
    /effects
    -0.06
    (Http
    -0.06
    _REG
    -0.06
     beating
    -0.06
    POSITIVE LOGITS
     �
    0.06
     Grocery
    0.06
    cpt
    0.06
     �
    0.06
    ACCEPT
    0.06
    bol
    0.06
    дает
    0.06
    véd
    0.06
     biking
    0.06
     �
    0.06
    Activations Density 0.015%

    No Known Activations