Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 92675
    Prev
    Next
    INDEX
    Explanations

    umbo

    np_max-act · gemini-2.0-flash

    high-resolution images depicting girls in various scenes and designs.

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

    This neuron detects high-level image-generation specification terms—like resolution (“4k,” “high-res”), quality/style adjectives (“masterpiece,” “cinematic lighting,” “finely detailed”), and camera or photographic references (“Hasselblad photography,” “dynamic angle,” “sharp focus”).

    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
     Hercules
    -0.06
    ğe
    -0.06
    暖
    -0.06
    нями
    -0.06
     shaving
    -0.06
    kám
    -0.06
    ースト
    -0.06
    B
    -0.06
    OVER
    -0.06
    امین
    -0.06
    POSITIVE LOGITS
    ومی
    0.07
    951
    0.07
     Bermuda
    0.06
    /DD
    0.06
     história
    0.06
     MUT
    0.06
     �
    0.06
    Restricted
    0.06
    <meta
    0.06
     jackets
    0.06
    Activations Density 0.026%

    No Known Activations