© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 5852
    Prev
    Next
    INDEX
    Explanations

    pornography

    np_max-act · gemini-2.0-flash

    The neuron fires on descriptive adjectives highlighting performers’ physical features—especially hairiness, body size (e.g. “big,” “chubby,” “natural”), and related attributes in pornographic content.

    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
    _yaml
    -0.07
     Sinclair
    -0.07
     Buddhist
    -0.06
     adultery
    -0.06
     damages
    -0.06
     optimal
    -0.06
     delete
    -0.06
     отли
    -0.06
    	image
    -0.06
     murdering
    -0.06
    POSITIVE LOGITS
    ุด
    0.06
    SED
    0.06
     Answer
    0.06
    íš
    0.06
    azz
    0.06
     racks
    0.06
    composed
    0.06
    �
    0.06
    \":{\"
    0.06
    监听
    0.06
    Activations Density 0.020%

    No Known Activations