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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 74668
    Prev
    Next
    INDEX
    Explanations

    U.S. patents

    np_max-act · gemini-2.0-flash

    The neuron selectively fires on tokens that are primarily numeric—e.g. the digit‐only segments you see in patent numbers, years, river lengths, and similar numeric identifiers.

    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
    Math
    -0.08
    mux
    -0.07
     Mend
    -0.07
    listen
    -0.07
     Peter
    -0.07
     ((
    -0.07
    Peter
    -0.06
     Kend
    -0.06
     Math
    -0.06
     Array
    -0.06
    POSITIVE LOGITS
    744
    0.09
    740
    0.08
    727
    0.08
    783
    0.08
    747
    0.08
    762
    0.08
    751
    0.08
    743
    0.08
    780
    0.08
    788
    0.08
    Activations Density 0.119%

    No Known Activations