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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 86136
    Prev
    Next
    INDEX
    Explanations

    Person's name and age

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting numeric tokens (e.g. ages, measurements, statistics) in the text.

    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
    Conversation
    -0.07
     الانت
    -0.06
     تقد
    -0.06
    자인
    -0.06
     müdür
    -0.06
    Cascade
    -0.06
    atum
    -0.06
    _digest
    -0.06
     trig
    -0.06
     racially
    -0.06
    POSITIVE LOGITS
    G
    0.07
     G
    0.06
     druhý
    0.06
    wik
    0.06
    inin
    0.06
    -old
    0.06
     Wit
    0.06
    RE
    0.06
     wit
    0.06
     Kı
    0.06
    Activations Density 0.023%

    No Known Activations