© 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. 74047
    Prev
    Next
    INDEX
    Explanations

    Extracting names or sports players

    np_max-act · gemini-2.0-flash

    This neuron detects placeholder tokens for named entities (e.g., “NAME_X”).

    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
    ,application
    -0.07
    ordes
    -0.06
     são
    -0.06
    RowAnimation
    -0.06
     тан
    -0.06
     Mev
    -0.06
     Smash
    -0.06
    Experimental
    -0.06
    Modify
    -0.06
     ה
    -0.06
    POSITIVE LOGITS
    ivity
    0.07
    =?,
    0.07
    _events
    0.06
     ilişk
    0.06
     назна
    0.06
     VARIABLES
    0.06
    重点
    0.06
    重複
    0.06
     CRE
    0.06
     подраз
    0.06
    Activations Density 0.017%

    No Known Activations