© 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. 90687
    Prev
    Next
    INDEX
    Explanations

    Places, sports categories

    np_max-act · gemini-2.0-flash

    The neuron fires on the noun or proper name that follows “in” in Wikipedia-style category lines (e.g. country names, demonyms, or domain topics in “Category:… in X”).

    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
    údo
    -0.07
    (fr
    -0.06
    odeled
    -0.06
    _switch
    -0.06
    _digest
    -0.06
    riteria
    -0.06
     Expo
    -0.06
     SON
    -0.06
    лов
    -0.06
     CONCAT
    -0.06
    POSITIVE LOGITS
     얼굴
    0.08
    силь
    0.07
    μένου
    0.07
     Disabilities
    0.07
     imagined
    0.07
    public
    0.07
     zb
    0.07
    36
    0.06
     consenting
    0.06
    ้เป
    0.06
    Activations Density 0.003%

    No Known Activations