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

    Neuronpedia

    Jacobian LensNEW
    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. 56769
    Prev
    Next
    INDEX
    Explanations

    quotations

    np_max-act · gemini-2.0-flash

    The neuron is activated by numeric tokens—particularly floating-point numbers (e.g. “0.388916015625,” “2.71875,” etc.)—and thus detects the presence of decimal numeric values.

    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
    -Qaeda
    -0.07
    ยา
    -0.06
    Dto
    -0.06
    	dto
    -0.06
    Magic
    -0.06
     pseud
    -0.06
    .persistent
    -0.06
    (pk
    -0.06
     Vend
    -0.06
    Ports
    -0.06
    POSITIVE LOGITS
     Expansion
    0.07
     setups
    0.07
    /community
    0.07
     arena
    0.07
     İl
    0.07
     expansion
    0.07
     climates
    0.06
    /Instruction
    0.06
     Gathering
    0.06
    .Direct
    0.06
    Activations Density 0.089%

    No Known Activations