© 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 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. 15-RESID-POST-AA
    11. 20611
    Prev
    Next
    INDEX
    Explanations

    debugging/logs

    np_max-act · gemini-2.0-flash

    The neuron detects special structural/control tokens and metadata boundaries (e.g., header/start/end markers and similar document-level tokens).

    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
    .collider
    -0.07
    界
    -0.07
    cec
    -0.06
    忙
    -0.06
     kitchens
    -0.06
    ,long
    -0.06
    кур
    -0.06
    زم
    -0.06
    فاق
    -0.06
    повід
    -0.06
    POSITIVE LOGITS
     ')↵
    0.08
     PLUS
    0.07
     unwilling
    0.06
    `↵↵
    0.06
     ');↵
    0.06
    /',↵
    0.06
    (description
    0.06
     Coll
    0.06
    .depth
    0.06
     Frem
    0.06
    Activations Density 0.532%

    No Known Activations