© 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. 41173
    Prev
    Next
    INDEX
    Explanations

    The neuron flags when the text is asking to set up or begin a “role play” scenario (e.g. mentions of “role,” “play,” “scenario,” “setting,” etc.).

    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
    ρών
    -0.07
     PATCH
    -0.07
    Approved
    -0.07
    Sm
    -0.07
    clamation
    -0.07
    •
    -0.07
     fluffy
    -0.06
    fullName
    -0.06
     FORE
    -0.06
     plaque
    -0.06
    POSITIVE LOGITS
    衝
    0.07
    568
    0.06
     сф
    0.06
    ницип
    0.06
     авт
    0.06
    บาล
    0.06
    !'↵
    0.06
    (SDL
    0.06
     důsled
    0.06
     DataColumn
    0.06
    Activations Density 0.017%

    No Known Activations