Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 42430
    Prev
    Next
    INDEX
    Explanations

    F

    np_max-act · gemini-2.0-flash

    This neuron fires on sequences of non-Latin script characters (e.g. Cyrillic, Greek, CJK) especially when the same character or block of such characters repeats.

    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
     مباش
    -0.07
    webElementXpaths
    -0.06
     Overrides
    -0.06
    ičky
    -0.06
    jylland
    -0.06
     Acceler
    -0.06
    leniyor
    -0.06
    quick
    -0.06
     ايران
    -0.06
    анню
    -0.06
    POSITIVE LOGITS
     sme
    0.07
    onChange
    0.06
     зак
    0.06
    �
    0.06
    奖
    0.06
     unbiased
    0.06
    /action
    0.06
     messy
    0.06
    ogonal
    0.06
     odp
    0.06
    Activations Density 0.002%

    No Known Activations