© 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. 91075
    Prev
    Next
    INDEX
    Explanations

    search queries and operators

    np_max-act · gemini-2.0-flash

    This neuron detects search‐operator tokens and their punctuation (e.g. “operators,” colons, plus/minus signs, parentheses, tildes) used to refine queries.

    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
     hatır
    -0.07
    CACHE
    -0.06
    타이
    -0.06
    .validator
    -0.06
    indir
    -0.06
     kino
    -0.06
     малень
    -0.06
    datap
    -0.06
    _lifetime
    -0.06
     Chile
    -0.06
    POSITIVE LOGITS
    izzes
    0.07
     willing
    0.06
    UnitTest
    0.06
     delet
    0.06
    Tags
    0.06
     meets
    0.06
    _prev
    0.06
     clarified
    0.06
    Issue
    0.06
     Adidas
    0.05
    Activations Density 0.007%

    No Known Activations