© 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. 30671
    Prev
    Next
    INDEX
    Explanations

    product

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the word “Product” (and its plural “products”), especially in section headings or product-description contexts.

    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
     gaze
    -0.08
    18
    -0.07
     fear
    -0.07
    43
    -0.07
     haste
    -0.07
    07
    -0.07
     Zone
    -0.06
    46
    -0.06
     sight
    -0.06
     Hole
    -0.06
    POSITIVE LOGITS
     product
    0.15
     products
    0.13
    product
    0.12
     Product
    0.12
    Product
    0.11
     Products
    0.11
    -product
    0.10
     productos
    0.10
    products
    0.09
    Products
    0.09
    Activations Density 0.080%

    No Known Activations