© 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. 7073
    Prev
    Next
    INDEX
    Explanations

    const

    np_max-act · gemini-2.0-flash

    The neuron detects occurrences of the JavaScript keyword “const” (i.e. block-scoped constant variable 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
    Ray
    -0.08
    Hay
    -0.07
    Example
    -0.07
    .web
    -0.07
     airplane
    -0.07
     Java
    -0.07
    Demo
    -0.07
     Ra
    -0.07
    .inv
    -0.07
    	KEY
    -0.07
    POSITIVE LOGITS
    0.07
     dismissing
    0.07
    š
    0.07
    σης
    0.07
    še
    0.07
    čen
    0.07
    ческим
    0.07
     ;
    0.06
     /
    0.06
    σεων
    0.06
    Activations Density 0.008%

    No Known Activations