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. 84536
    Prev
    Next
    INDEX
    Explanations

    martial arts

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of specific martial-arts styles, schools, or systems (e.g. “Lameco Eskrima,” “Aikido,” “Brazilian Jiu-Jitsu,” “Taekwondo,” etc.).

    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
     regression
    -0.07
     gunfire
    -0.07
     spheres
    -0.06
     گفته
    -0.06
    马
    -0.06
    _ts
    -0.06
    ='<
    -0.06
     Fecha
    -0.06
     FIXME
    -0.06
    combine
    -0.06
    POSITIVE LOGITS
    문을
    0.07
     disputes
    0.06
     Robin
    0.06
     دول
    0.06
     způsob
    0.06
    ेम
    0.06
    .getvalue
    0.06
    .TypeOf
    0.06
    /url
    0.06
     diversion
    0.06
    Activations Density 0.016%

    No Known Activations