© 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. 29785
    Prev
    Next
    INDEX
    Explanations

    Q and Hall

    np_max-act · gemini-2.0-flash

    This neuron detects chess move tokens, specifically uppercase piece‐notation symbols (like B, N, R, Q) and their compound forms in algebraic notation.

    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
    ден
    -0.07
     zoals
    -0.07
     prá
    -0.07
     CAT
    -0.07
    реб
    -0.06
     пацієн
    -0.06
     sane
    -0.06
     pertinent
    -0.06
    114
    -0.06
     Spare
    -0.06
    POSITIVE LOGITS
     hallway
    0.07
     Hall
    0.06
     cuisine
    0.06
     '"'
    0.06
     ancestral
    0.06
    ublished
    0.06
    .bounds
    0.06
    .Side
    0.06
     hall
    0.06
    Board
    0.06
    Activations Density 0.001%

    No Known Activations