© 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. 110815
    Prev
    Next
    INDEX
    Explanations

    entanglement

    np_max-act · gemini-2.0-flash

    This neuron selectively responds to explicit sexual or erotic action terms, such as “kiss,” “fucks,” “French,” “slobbery,” and similar sexual-activity words.

    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
     เจ
    -0.08
     QStringLiteral
    -0.07
    788
    -0.07
    >Loading
    -0.06
    .Serve
    -0.06
     يا
    -0.06
    Seeing
    -0.06
     Před
    -0.06
     dark
    -0.06
     طلب
    -0.06
    POSITIVE LOGITS
     rámci
    0.07
     buys
    0.06
     Stat
    0.06
     impr
    0.06
     LOOP
    0.06
     presses
    0.06
     Mentor
    0.06
    AMILY
    0.06
     migrant
    0.06
     brutally
    0.06
    Activations Density 0.002%

    No Known Activations