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. 130790
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron detects technical descriptions of geometric structures in mathematical contexts, especially phrases introducing a group (orbits) equipped with an invariant metric.

    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
     Emotional
    -0.06
     midfielder
    -0.06
     television
    -0.06
    (Board
    -0.06
     Numbers
    -0.06
     lyrics
    -0.06
     psyche
    -0.06
    .EntityFramework
    -0.06
     Nun
    -0.06
    Elite
    -0.06
    POSITIVE LOGITS
     사업
    0.07
     신청
    0.06
    semblies
    0.06
     获取
    0.06
    .decorate
    0.06
     lassen
    0.06
    ':↵
    0.06
    闪
    0.06
     тех
    0.06
     çözüm
    0.06
    Activations Density 0.042%

    No Known Activations