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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 43536
    Prev
    Next
    INDEX
    Explanations

    j

    np_max-act · gemini-2.0-flash

    The neuron flags tokens that refer to XML‐ and JAXB‐related identifiers—e.g. parts of “javax.xml.bind,” XML namespace URLs, and similar XML/JAXB class or schema names.

    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
    '];↵
    -0.07
    (L
    -0.07
    わず
    -0.06
    (Unknown
    -0.06
    ┃
    -0.06
    +"\
    -0.06
    ';↵↵
    -0.06
     jade
    -0.06
    ,password
    -0.06
    (indent
    -0.06
    POSITIVE LOGITS
     зрения
    0.06
     Welt
    0.06
    ์ส
    0.06
     Derived
    0.06
    turnstile
    0.06
     spun
    0.06
     Merch
    0.06
    Installed
    0.06
     tph
    0.06
     Vers
    0.06
    Activations Density 0.022%

    No Known Activations