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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 58898
    Prev
    Next
    INDEX
    Explanations

    place

    np_max-act · gemini-2.0-flash

    The neuron is a detector for the word “Workplace” (i.e. the token sequence spelling “Workplace” or “workplace”).

    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
     omit
    -0.07
     shows
    -0.07
     McCain
    -0.07
     kilometers
    -0.07
     traveling
    -0.06
     signer
    -0.06
    262
    -0.06
     clear
    -0.06
     honors
    -0.06
    itus
    -0.06
    POSITIVE LOGITS
     workplace
    0.12
     Workplace
    0.11
     marketplace
    0.09
    ах
    0.08
     Marketplace
    0.08
    pecia
    0.08
     workforce
    0.07
    페
    0.07
     workplaces
    0.07
    γε
    0.07
    Activations Density 0.008%

    No Known Activations