© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 128752
    Prev
    Next
    INDEX
    Explanations

    netty

    np_max-act · gemini-2.0-flash

    The neuron fires on the “ty” token in package and class names—i.e. it’s detecting mentions of the Netty/Jetty networking libraries.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    code-related elements, particularly method definitions and annotations in programming context.

    oai_token-act-pair · gpt-4o-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
     fois
    -0.07
    unts
    -0.06
     Dra
    -0.06
     이상
    -0.06
     Shader
    -0.06
    Door
    -0.06
    605
    -0.06
    endance
    -0.06
    ​​
    -0.06
     ventilation
    -0.06
    POSITIVE LOGITS
    ρω
    0.08
    ty
    0.07
    ewriter
    0.07
     Ли
    0.07
    $path
    0.06
     Guzzle
    0.06
     лікар
    0.06
    ycastle
    0.06
     Petty
    0.06
     ConcurrentHashMap
    0.06
    Activations Density 0.001%

    No Known Activations