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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 91402
    Prev
    Next
    INDEX
    Explanations

    -instance

    np_max-act · gemini-2.0-flash

    This neuron detects the XML Schema‐instance namespace declaration (e.g. the “http://www.w3.org/2001/XMLSchema-instance” used in POM files).

    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
     rotten
    -0.07
    humidity
    -0.06
    Starting
    -0.06
    bstract
    -0.06
     defend
    -0.06
     Moran
    -0.06
    health
    -0.06
    _platform
    -0.06
     WC
    -0.06
     Develop
    -0.06
    POSITIVE LOGITS
    mini
    0.07
    nett
    0.07
    اعية
    0.07
    "></
    0.06
    -do
    0.06
    -ever
    0.06
     ".$
    0.06
     invers
    0.06
     ORDER
    0.06
     confirmation
    0.06
    Activations Density 0.000%

    No Known Activations