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

    Neuronpedia

    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. 11-RESID-POST-AA
    11. 9123
    Prev
    Next
    INDEX
    Explanations

    programming code strings

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on Swift function or initializer calls that include labeled parameters in parentheses (e.g. URL(string:…), Data(contentsOf:…), JSONDecoder.decode(_:from:), etc.).

    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
    venience
    -0.07
    _spacing
    -0.06
    _salary
    -0.06
     dzi
    -0.06
    acı
    -0.06
     Mac
    -0.06
    _DISTANCE
    -0.06
    .
↵
↵
    -0.06
    .t
    -0.06
    Shoot
    -0.06
    POSITIVE LOGITS
    解
    0.08
    nard
    0.07
     flashback
    0.07
    @RestController
    0.06
    [selected
    0.06
    /operators
    0.06
     أمريكي
    0.06
     northeastern
    0.06
     friendships
    0.06
    chn
    0.06
    Activations Density 0.002%

    No Known Activations