© 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. 43269
    Prev
    Next
    INDEX
    Explanations

    morality

    np_max-act · gemini-2.0-flash

    This neuron detects tokens related to moral transgression or the absence of ethical constraints (e.g. “immoral,” “unethical,” “amoral,” “Machiavellian”).

    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
     AK
    -0.06
     eased
    -0.06
     áp
    -0.06
    执
    -0.06
     hely
    -0.06
     giá
    -0.06
     podí
    -0.06
    ��
    -0.06
     گذشته
    -0.06
    .Ad
    -0.06
    POSITIVE LOGITS
    원을
    0.07
    .Files
    0.07
    .Time
    0.07
     Confirm
    0.07
     CHR
    0.06
    inst
    0.06
    	Object
    0.06
    روت
    0.06
     Tham
    0.06
    Processes
    0.06
    Activations Density 0.009%

    No Known Activations