© 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. 72212
    Prev
    Next
    INDEX
    Explanations

    Portuguese

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on subword tokens that belong to Portuguese (e.g. roots and inflectional pieces like “tent,” “mens,” “iça,” “ficheiro,” etc.), effectively flagging Portuguese-language content.

    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
     Kraft
    -0.07
    wstring
    -0.07
     Gef
    -0.06
    St
    -0.06
     Ward
    -0.06
     द
    -0.06
     pose
    -0.06
    (number
    -0.06
     Temple
    -0.06
    와
    -0.06
    POSITIVE LOGITS
    abela
    0.07
     "),↵
    0.07
     короб
    0.06
     Lisbon
    0.06
    ')
↵
↵
    0.06
     باشید
    0.06
    уванням
    0.06
     tipos
    0.06
    ками
    0.06
    ocê
    0.06
    Activations Density 0.085%

    No Known Activations