© 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 EvalsExportsinterp-engineNEWAPI 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. 58316
    Prev
    Next
    INDEX
    Explanations

    numbered lists/instructions

    np_max-act · gemini-2.0-flash

    The neuron primarily detects numbered‐step markers (e.g. “1)”, “2)”, etc.) in the instruction lists.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    instructions related to Python coding and variable management.

    oai_token-act-pair · gpt-4o-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
     ابتد
    -0.07
     平方
    -0.07
     Flood
    -0.06
     Preconditions
    -0.06
     Scot
    -0.06
    (command
    -0.06
    basePath
    -0.06
     orada
    -0.06
     Hyp
    -0.06
     пораж
    -0.06
    POSITIVE LOGITS
     conoc
    0.07
    ManagedObject
    0.07
     качества
    0.07
    ственным
    0.07
     prze
    0.06
    ्तक
    0.06
    .Listen
    0.06
    >".$
    0.06
     proceedings
    0.06
     metabolism
    0.06
    Activations Density 0.029%

    No Known Activations