© 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. 124830
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron detects imperative action phrases—commands telling you to do something (e.g. “Delete the derived data,” “Clean the project,” “Restart the OS”).

    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
     summaries
    -0.07
     sama
    -0.07
    msg
    -0.07
    893
    -0.06
    emia
    -0.06
    (spell
    -0.06
    (vector
    -0.06
    arna
    -0.06
    ){↵
    -0.06
    dst
    -0.06
    POSITIVE LOGITS
    рол
    0.07
    Segue
    0.07
    PointF
    0.06
     SCE
    0.06
     promo
    0.06
     ATH
    0.06
    RoutingModule
    0.06
    ển
    0.06
    _ENC
    0.06
    ่เป
    0.06
    Activations Density 0.015%

    No Known Activations