© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 74031
    Prev
    Next
    INDEX
    Explanations

    Legal and official documents

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on numerical and alphanumeric references typical of legal citations or codes (e.g. section/article numbers, reporter volumes, dates, phone numbers).

    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
     shade
    -0.07
    .instant
    -0.06
     rune
    -0.06
     शर
    -0.06
    })();
    -0.06
    hammad
    -0.06
     ува
    -0.06
    "W
    -0.06
     UserService
    -0.06
    -imm
    -0.06
    POSITIVE LOGITS
    ersist
    0.07
     происходит
    0.07
    OCUMENT
    0.06
    _validate
    0.06
    owner
    0.06
     VIR
    0.06
     rename
    0.06
     Ptr
    0.06
    assoc
    0.06
    .Dynamic
    0.06
    Activations Density 0.155%

    No Known Activations