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. 130487
    Prev
    Next
    INDEX
    Explanations

    auxiliary verbs

    np_max-act · gemini-2.0-flash

    This neuron detects words that signal detailed exploration or technical methodology, such as “delves” and “decoding.”

    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
    .Cross
    -0.07
    _qty
    -0.07
    –and
    -0.07
     and
    -0.06
    Booking
    -0.06
     AND
    -0.06
     ژاپ
    -0.06
    lightbox
    -0.06
    .baseUrl
    -0.06
     двиг
    -0.06
    POSITIVE LOGITS
    awk
    0.08
     civilian
    0.06
     atl
    0.06
     이
    0.06
     CORPORATION
    0.06
    ský
    0.06
     cái
    0.06
    州
    0.06
     obesity
    0.06
     yön
    0.06
    Activations Density 0.245%

    No Known Activations