© 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. 30887
    Prev
    Next
    INDEX
    Explanations

    with

    np_max-act · gemini-2.0-flash

    The neuron specifically detects the instruction phrase that begins the draft—namely the “You start off with “ marker.

    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
    １
    -0.06
    Plus
    -0.06
    _successful
    -0.06
     após
    -0.06
     director
    -0.06
    '])){↵
    -0.06
    寄
    -0.06
     Positions
    -0.06
     subTitle
    -0.06
    �
    -0.06
    POSITIVE LOGITS
     LIMITED
    0.06
    ист
    0.06
     DISP
    0.06
     دوباره
    0.06
     hesap
    0.06
    िज
    0.06
     rámci
    0.06
    hait
    0.06
     frenzy
    0.06
    辰
    0.06
    Activations Density 0.001%

    No Known Activations