© 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. 129682
    Prev
    Next
    INDEX
    Explanations

    domain endings

    np_max-act · gemini-2.0-flash

    The neuron identifies top‐level domain or organization names (e.g. “com,” “org,” “io,” “net”) that appear as the first segment of Java package or fully qualified class 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
    ря
    -0.07
     servlet
    -0.07
    /angular
    -0.07
     TEM
    -0.06
     Compet
    -0.06
    	n
    -0.06
    -pages
    -0.06
    TEM
    -0.06
    scription
    -0.06
    ustainable
    -0.06
    POSITIVE LOGITS
    akespeare
    0.07
     DateTimeOffset
    0.07
     шп
    0.07
    _bar
    0.06
     √
    0.06
     Shakespeare
    0.06
    Ô
    0.06
    �
    0.06
     dollars
    0.06
    spell
    0.06
    Activations Density 0.001%

    No Known Activations