© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 92334
    Prev
    Next
    INDEX
    Explanations

    news articles

    np_max-act · gemini-2.0-flash

    This neuron fires on the typical lead‐in phrases and formatting cues of a news article—especially the opening dateline or paragraph-starter expressions (e.g. “In a shocking turn of events,” “According to sources,” the quoted headline)—marking the start of new sections in a journalistic write‐up.

    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
     amazed
    -0.07
     Myanmar
    -0.06
    習
    -0.06
    uç
    -0.06
    stanov
    -0.06
    WISE
    -0.06
     الرئيس
    -0.06
     آنان
    -0.06
    lara
    -0.06
     vzdu
    -0.06
    POSITIVE LOGITS
     visionary
    0.07
     ulus
    0.07
     erro
    0.07
     depressive
    0.07
     Branch
    0.06
     preco
    0.06
         ↵
    0.06
     ем
    0.06
    .Screen
    0.06
     epith
    0.06
    Activations Density 0.027%

    No Known Activations