© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 12630
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates on JavaScript “window == top” (or “window === top”) checks—that is, it spots code testing whether the page is running inside an iframe.

    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
    stor
    -0.07
     ocur
    -0.07
    Declaration
    -0.06
     beet
    -0.06
     Dok
    -0.06
    DRAM
    -0.06
     TWO
    -0.06
    オリ
    -0.06
    622
    -0.06
    Transformation
    -0.06
    POSITIVE LOGITS
    coords
    0.08
     verde
    0.07
    <Action
    0.07
     запах
    0.06
     hybrid
    0.06
    _submenu
    0.06
    ("<?
    0.06
    [jj
    0.06
    .RegisterType
    0.06
    alien
    0.06
    Activations Density 0.000%

    No Known Activations