© 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. 109251
    Prev
    Next
    INDEX
    Explanations

    code snippets

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on CamelCase identifiers (like class names or file/path tokens), essentially flagging code‐style names in the text.

    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
     Shoes
    -0.07
     Hyundai
    -0.06
     Ý
    -0.06
     memoria
    -0.06
    -spin
    -0.06
    UY
    -0.06
    .getBean
    -0.06
     robots
    -0.06
     opera
    -0.06
    ῖ
    -0.06
    POSITIVE LOGITS
     مشاهده
    0.07
    ={(
    0.07
     figur
    0.06
     ActionTypes
    0.06
    Filed
    0.06
    (images
    0.06
     شاید
    0.06
    NavigationView
    0.06
     persisted
    0.06
     معت
    0.06
    Activations Density 0.031%

    No Known Activations