© 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. 54499
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    This neuron detects section‐or report‐structure markers, specifically phrases like “first part of” and “second part of” that outline the document’s organization.

    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
    Fi
    -0.06
     Future
    -0.06
    _xx
    -0.06
     wreck
    -0.06
    /temp
    -0.06
     weg
    -0.06
     Kids
    -0.06
     Season
    -0.06
     kids
    -0.05
     PLAN
    -0.05
    POSITIVE LOGITS
     arguably
    0.06
     colonization
    0.06
    covers
    0.06
     *)↵↵
    0.06
    Outlet
    0.06
     relics
    0.06
    _in
    0.06
    _keys
    0.06
     ainsi
    0.06
    .mongodb
    0.06
    Activations Density 0.060%

    No Known Activations