Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 8290
    Prev
    Next
    INDEX
    Explanations

    <

    np_max-act · gemini-2.0-flash

    The neuron fires on the angle-bracketed protocol conformance clauses (e.g. “<NSCopying>”, “<RCTBridgeModule>”, etc.) in Objective-C @interface declarations.

    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
    =os
    -0.06
    ONGO
    -0.06
     fileprivate
    -0.06
     hastalık
    -0.06
     Granny
    -0.06
     تبد
    -0.06
    "))
    -0.06
    -alist
    -0.06
    다
    -0.06
    uger
    -0.06
    POSITIVE LOGITS
     Pret
    0.07
    _fore
    0.07
     bleak
    0.07
    virtual
    0.06
    �
    0.06
     Lorem
    0.06
     Shape
    0.06
    incre
    0.06
     PRICE
    0.06
    _Panel
    0.06
    Activations Density 0.001%

    No Known Activations