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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126103
    Prev
    Next
    INDEX
    Explanations

    Latin or Greek

    np_max-act · gemini-2.0-flash

    This neuron activates on mentions of source languages (e.g. “Latin,” “French”) in etymological explanations.

    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
    —I
    -0.07
    最近
    -0.07
    φων
    -0.06
    ステム
    -0.06
     '/',
    -0.06
    ров
    -0.06
     overflowing
    -0.06
     fourteen
    -0.06
     Đây
    -0.06
    股份有限公司
    -0.06
    POSITIVE LOGITS
     mak
    0.06
     funk
    0.06
    reserved
    0.06
    ifa
    0.06
     persisted
    0.06
    (accounts
    0.06
     bounty
    0.06
     Ukra
    0.06
     staggering
    0.06
    мерик
    0.06
    Activations Density 0.008%

    No Known Activations