© 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. 31053
    Prev
    Next
    INDEX
    Explanations

    Korean entertainment news

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on named entities and titles—that is, capitalized words and proper names (e.g. show titles, person names, organization names).

    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
    sterreich
    -0.07
     любов
    -0.07
     cov
    -0.06
     Provid
    -0.06
     parm
    -0.06
     सह
    -0.06
     yapıl
    -0.06
    >M
    -0.06
     BU
    -0.06
     ').
    -0.06
    POSITIVE LOGITS
     guilty
    0.07
    .setPreferredSize
    0.07
    ज
    0.07
    (padding
    0.07
    charging
    0.06
    има
    0.06
    REPORT
    0.06
    ęp
    0.06
    STATIC
    0.06
     конт
    0.06
    Activations Density 0.013%

    No Known Activations