© 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. 15-RESID-POST-AA
    11. 45203
    Prev
    Next
    INDEX
    Explanations

    chatbots and computers

    np_max-act · gemini-2.0-flash

    tokens and short phrases that appear in system prompts, role labels, or persona-setting/instructional context (e.g., role names and directive words).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Walking
    -0.07
    eroon
    -0.07
     yemek
    -0.06
    'an
    -0.06
    _vid
    -0.06
     guarante
    -0.06
    unami
    -0.06
    mtime
    -0.06
    ViewPager
    -0.06
    .App
    -0.06
    POSITIVE LOGITS
    .Raw
    0.07
    .asset
    0.07
     #↵
    0.06
     prototypes
    0.06
    inic
    0.06
    (latitude
    0.06
    (false
    0.06
    (ALOAD
    0.06
     Giants
    0.06
     rgb
    0.06
    Activations Density 0.065%

    No Known Activations