© 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. 130253
    Prev
    Next
    INDEX
    Explanations

    match

    np_max-act · gemini-2.0-flash

    The neuron is detecting numeric literal tokens—particularly floating‐point numbers (e.g., 3.99, 2.40) in XML layout attribute values like layout_width and layout_height.

    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
     bites
    -0.08
     yacht
    -0.06
     Ride
    -0.06
    -runner
    -0.06
    -shift
    -0.06
    егод
    -0.06
     alunos
    -0.06
     sd
    -0.06
     JD
    -0.06
    /false
    -0.06
    POSITIVE LOGITS
        	
↵
    0.07
    ㅋㅋ
    0.07
     виб
    0.06
    								
↵
    0.06
     clientele
    0.06
    _marker
    0.06
     pov
    0.06
     필
    0.06
     piyas
    0.06
     vysvět
    0.06
    Activations Density 0.001%

    No Known Activations