© 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. 101832
    Prev
    Next
    INDEX
    Explanations

    Questions and answers

    np_max-act · gemini-2.0-flash

    the neuron detects words and tokens that signal generality or broad scope in instructions (e.g., "any", "any type", and similar markers of open/broad requests).

    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_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
    Titles
    -0.06
     Museum
    -0.06
    party
    -0.06
     casual
    -0.06
    門
    -0.06
    directory
    -0.06
     lapse
    -0.06
    teachers
    -0.06
    ادم
    -0.06
    	port
    -0.06
    POSITIVE LOGITS
     Romanian
    0.07
    (I
    0.07
    OLUM
    0.07
     �
    0.07
    .UserName
    0.07
    ]");↵
    0.06
     Farage
    0.06
    віт
    0.06
    沖
    0.06
     нес
    0.06
    Activations Density 0.234%

    No Known Activations