© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 129597
    Prev
    Next
    INDEX
    Explanations

    prejudice

    np_max-act · gemini-2.0-flash

    references to social issues related to discrimination and bias.

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

    This neuron activates on terms describing social bias and unfair treatment—words like “prejudice,” “discrimination,” and “racism.”

    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
    _roll
    -0.07
     sacks
    -0.07
    FontSize
    -0.07
    alement
    -0.06
     wlan
    -0.06
     Lori
    -0.06
    .strftime
    -0.06
     Automation
    -0.06
    CardContent
    -0.06
     автомати
    -0.06
    POSITIVE LOGITS
     prejudice
    0.12
     prejudices
    0.11
    рот
    0.07
    -Th
    0.07
     superst
    0.07
     producing
    0.07
     prejud
    0.07
    quil
    0.06
    uesta
    0.06
     influences
    0.06
    Activations Density 0.005%

    No Known Activations