© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 59373
    Prev
    Next
    INDEX
    Explanations

    image

    np_max-act · gemini-2.0-flash

    metadata tags used in HTML documents, specifically for social media sharing attributes.

    oai_token-act-pair · gpt-4o-mini
    Triggered by @xinyanhu8

    The neuron is detecting HTML meta tags that declare social‐media “og:” (Open Graph) properties (e.g. og:title, og:image, og:url).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    .cap
    -0.07
    ότητας
    -0.07
    atori
    -0.06
    akhir
    -0.06
     Shutterstock
    -0.06
     appBar
    -0.06
     đông
    -0.06
     Payne
    -0.06
     chipset
    -0.06
    agh
    -0.06
    POSITIVE LOGITS
     Michelle
    0.07
     WebElement
    0.07
     ],↵
    0.07
    Michelle
    0.06
     ↵↵
    0.06
    t
    0.06
    도가
    0.06
     zbyt
    0.06
    .Symbol
    0.06
    getImage
    0.06
    Activations Density 0.004%

    No Known Activations