© 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. 42901
    Prev
    Next
    INDEX
    Explanations

    hair

    np_max-act · gemini-2.0-flash

    The neuron is sensitive to explicit anatomical and erotic body‐part terms, especially references to pubic hair and nearby sexual regions.

    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
     eventId
    -0.06
     skull
    -0.06
     �
    -0.06
    FindObject
    -0.06
     horr
    -0.06
     Damon
    -0.06
    .firebaseapp
    -0.06
    уть
    -0.06
    Mix
    -0.06
    (bounds
    -0.06
    POSITIVE LOGITS
    coil
    0.07
    υνα
    0.07
    คล
    0.06
    	InputStream
    0.06
    	public
    0.06
    243
    0.06
                            ↵
    0.06
    	username
    0.06
     tranh
    0.06
     ngờ
    0.06
    Activations Density 0.004%

    No Known Activations