© 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. 12354
    Prev
    Next
    INDEX
    Explanations

    which

    np_max-act · gemini-2.0-flash

    This neuron responds to subordinating words (like “that,” “which,” and “is”) that introduce or link clauses in technical descriptions.

    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
     famous
    -0.07
    (cm
    -0.07
     merely
    -0.06
     Homer
    -0.06
     qreal
    -0.06
     coverage
    -0.06
    {},
    -0.06
    文化
    -0.06
    .save
    -0.06
     fitted
    -0.06
    POSITIVE LOGITS
    ází
    0.07
     hôm
    0.06
    @
    0.06
    ={}
    0.06
    명의
    0.06
     Moines
    0.06
    orghini
    0.06
     Dual
    0.06
    ashire
    0.06
    ession
    0.06
    Activations Density 0.057%

    No Known Activations