© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126226
    Prev
    Next
    INDEX
    Explanations

    Safety and health

    np_max-act · gemini-2.0-flash

    content related to illegal or harmful sexual activities.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron detects tokens used in moral‐safety advice—especially words like “prioritize,” “safety,” and “well-being.”

    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
    .COLOR
    -0.07
     orthodox
    -0.06
    CLOCKS
    -0.06
    adapt
    -0.06
    گوی
    -0.06
    elor
    -0.06
    loating
    -0.06
    esiz
    -0.06
    xpath
    -0.06
    ῶ
    -0.06
    POSITIVE LOGITS
     Links
    0.07
     Guest
    0.07
     hitch
    0.07
     park
    0.06
    ARDS
    0.06
     extr
    0.06
     movements
    0.06
     loyalty
    0.06
     know
    0.06
     death
    0.06
    Activations Density 0.022%

    No Known Activations