© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109242
    Prev
    Next
    INDEX
    Explanations

    Es

    np_max-act · gemini-2.0-flash

    The neuron activates on attorney and counsel names (e.g. “Esquire,” individual lawyer names, and law‐firm attribution lines).

    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
     niños
    -0.07
     réuss
    -0.06
    ryfall
    -0.06
    edral
    -0.06
    erman
    -0.06
     kennen
    -0.06
    ()){
↵
    -0.06
    ابط
    -0.06
    ANEL
    -0.06
    ﺔ
    -0.06
    POSITIVE LOGITS
    /or
    0.07
     @"";↵
    0.06
     unaffected
    0.06
    _STATE
    0.06
    (mut
    0.06
    ‌س
    0.06
     امنیت
    0.06
    ",$
    0.06
     Cleans
    0.06
     bist
    0.06
    Activations Density 0.005%

    No Known Activations