Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 84635
    Prev
    Next
    INDEX
    Explanations

    news articles

    np_max-act · gemini-2.0-flash

    This neuron fires on uppercase or headline‐style tokens (e.g. all-caps place names, section headings, and error/exception names).

    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
    .datasource
    -0.06
    етод
    -0.06
     servidor
    -0.06
    sparse
    -0.06
    EDI
    -0.06
    ocus
    -0.06
     his
    -0.06
    .exp
    -0.06
     sound
    -0.06
    	action
    -0.06
    POSITIVE LOGITS
     avons
    0.08
    体系
    0.07
    .djangoproject
    0.07
     Flexible
    0.06
     Tar
    0.06
     Lunar
    0.06
    ;.
    0.06
     프
    0.06
    olute
    0.06
     ΑΝ
    0.06
    Activations Density 0.138%

    No Known Activations