© 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 EvalsExportsinterp-engineNEWAPI 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. 58876
    Prev
    Next
    INDEX
    Explanations

    cultural exchange

    np_max-act · gemini-2.0-flash

    The neuron selectively fires on hyphenated compounds that denote cross‐regional or cross‐cultural connections (e.g. “cross-cultural,” “trans-Atlantic,” “trans-continental”).

    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
    yal
    -0.07
    velte
    -0.07
     ket
    -0.07
    Jer
    -0.07
    .).
    -0.06
    ,copy
    -0.06
     Jer
    -0.06
    _adjust
    -0.06
     giz
    -0.06
     voice
    -0.06
    POSITIVE LOGITS
     epile
    0.06
    する
    0.06
     textured
    0.06
     pav
    0.06
     사업
    0.06
     LoggerFactory
    0.06
    促
    0.06
    %=
    0.06
    (Login
    0.06
     Patent
    0.06
    Activations Density 0.014%

    No Known Activations