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. 86202
    Prev
    Next
    INDEX
    Explanations

    mathematical equations

    np_max-act · gemini-2.0-flash

    This neuron activates strongly on specialized gauge‐theory terminology—especially occurrences of “Chern-Simons,” “gauge field,” “connection,” 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
    iband
    -0.07
     parade
    -0.06
     Trainer
    -0.06
     interpersonal
    -0.06
    的に
    -0.06
    buffers
    -0.06
     IPL
    -0.06
    (Transform
    -0.06
    Fx
    -0.06
     isize
    -0.06
    POSITIVE LOGITS
     воды
    0.07
    rang
    0.07
    ang
    0.06
     Publisher
    0.06
     Producto
    0.06
    product
    0.06
     arranging
    0.06
     Arithmetic
    0.06
    �
    0.06
    Producto
    0.06
    Activations Density 0.002%

    No Known Activations