© 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. 126741
    Prev
    Next
    INDEX
    Explanations

    politics

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of political discourse, especially the words “politics” and “political.”

    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
    каз
    -0.08
     glyph
    -0.07
    독
    -0.07
    кат
    -0.06
     درخواست
    -0.06
    캐
    -0.06
    Sync
    -0.06
     RAF
    -0.06
     Coch
    -0.06
     Jo
    -0.06
    POSITIVE LOGITS
    _Product
    0.06
    _public
    0.06
    .Errors
    0.06
    üslü
    0.06
    ัณฑ
    0.06
    constant
    0.06
     čím
    0.06
    flate
    0.06
    forg
    0.06
    -lived
    0.06
    Activations Density 0.009%

    No Known Activations