© 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. 13764
    Prev
    Next
    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash

    The neuron activates on bits of legal citation text—case names, statutory references, “see,” “People v. …,” section numbers, and related citation tokens.

    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
     Arizona
    -0.06
     wasm
    -0.06
    σμα
    -0.06
    �
    -0.06
    ILT
    -0.06
    [Math
    -0.06
    Vo
    -0.06
    ckill
    -0.06
     Suriye
    -0.06
    =url
    -0.06
    POSITIVE LOGITS
    تا
    0.06
    ufficient
    0.06
    _St
    0.06
     Township
    0.06
     sights
    0.06
    sequence
    0.06
     cutting
    0.06
     wedding
    0.06
     presidency
    0.06
     trình
    0.06
    Activations Density 0.002%

    No Known Activations