© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 38693
    Prev
    Next
    INDEX
    Explanations

    when

    np_max-act · gemini-2.0-flash

    This neuron detects discourse markers—especially subordinating conjunctions or adverbial connectors (e.g., “when,” “especially”) that introduce conditions or emphasis.

    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
     kullan
    -0.07
     etkin
    -0.06
    ]["
    -0.06
    ى
    -0.06
     مست
    -0.06
     scouting
    -0.06
    拉
    -0.06
    溫
    -0.06
    ‌سی
    -0.06
     대전
    -0.06
    POSITIVE LOGITS
     Constitution
    0.07
    Understanding
    0.07
     захворю
    0.07
    psy
    0.07
     LIABILITY
    0.06
    angible
    0.06
    angles
    0.06
    bral
    0.06
     Goods
    0.06
    Feb
    0.06
    Activations Density 0.030%

    No Known Activations