© 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. 6376
    Prev
    Next
    INDEX
    Explanations

    ad

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on the substring “ad” (as in “ad,” “advertising,” etc.), effectively detecting advertisement-related tokens.

    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
     ")[
    -0.06
    Rand
    -0.06
    Arduino
    -0.06
    done
    -0.06
    刚才
    -0.06
    mousedown
    -0.06
    ;y
    -0.06
    「我
    -0.06
     continua
    -0.06
    Ly
    -0.05
    POSITIVE LOGITS
     sexism
    0.08
     Binding
    0.07
     fotos
    0.07
    _added
    0.07
    sole
    0.07
    .area
    0.06
     Türk
    0.06
     John
    0.06
    -Se
    0.06
     tổn
    0.06
    Activations Density 0.041%

    No Known Activations