© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 41242
    Prev
    Next
    INDEX
    Explanations

    +

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting LaTeX-style superscript and exponent notation (the caret “^” and its following plus/minus signs and digits) in mathematical expressions.

    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
     reefs
    -0.07
    {id
    -0.06
    istence
    -0.06
    qrstuvwxyz
    -0.06
    ominated
    -0.06
     s
    -0.05
     نو
    -0.05
    ennis
    -0.05
     Strength
    -0.05
    Jur
    -0.05
    POSITIVE LOGITS
     product
    0.07
     supply
    0.07
     reside
    0.07
    Publish
    0.07
    packed
    0.07
     Croatia
    0.07
     Debt
    0.07
     Serum
    0.06
    ща
    0.06
    ませ
    0.06
    Activations Density 0.001%

    No Known Activations