Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 91103
    Prev
    Next
    INDEX
    Explanations

    Track lists for musicians

    np_max-act · gemini-2.0-flash

    The neuron fires on numeric tokens inside parentheses—i.e. the parenthesized track numbers and related numeric lists in the album’s credits.

    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
     Surveillance
    -0.07
    paired
    -0.06
    네
    -0.06
    neapolis
    -0.06
    _ur
    -0.06
     DropIndex
    -0.06
    ModuleName
    -0.06
    ावन
    -0.06
    ۱۹۹
    -0.06
     людини
    -0.06
    POSITIVE LOGITS
     Bever
    0.08
    (card
    0.07
     CAN
    0.07
     cush
    0.06
    	sm
    0.06
     Cah
    0.06
     Sioux
    0.06
    .req
    0.06
     ipAddress
    0.06
     Καρ
    0.06
    Activations Density 0.003%

    No Known Activations