© 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. 28092
    Prev
    Next
    INDEX
    Explanations

    questions and answers

    np_max-act · gemini-2.0-flash

    the assistant turn marker—the tokens indicating the start of an assistant’s response in a chat-formatted transcript.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
    .boot
    -0.07
     agora
    -0.07
     Khoa
    -0.06
     Flex
    -0.06
    <th
    -0.06
     Baxter
    -0.06
    護
    -0.06
    ným
    -0.06
    ximo
    -0.06
    GOR
    -0.06
    POSITIVE LOGITS
     Ark
    0.06
    Detalle
    0.06
     fantasy
    0.06
     ATA
    0.06
     parte
    0.06
    ACH
    0.06
    Located
    0.06
     communion
    0.06
    _female
    0.06
    雙
    0.06
    Activations Density 0.039%

    No Known Activations