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. 130346
    Prev
    Next
    INDEX
    Explanations

    colon

    np_max-act · gemini-2.0-flash

    The neuron detects Sphinx/reStructuredText autodoc directive lines (e.g. “.. automodule::” or “.. autoclass::”) in documentation.

    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
     그녀는
    -0.07
    ग
    -0.06
     сид
    -0.06
     Miz
    -0.06
     Bamboo
    -0.06
     başına
    -0.06
    商品
    -0.06
     cheese
    -0.06
    -clear
    -0.06
    ▲
    -0.06
    POSITIVE LOGITS
    κτη
    0.07
     tedious
    0.07
    ngine
    0.07
    authorize
    0.07
     statuses
    0.06
    خیص
    0.06
    uly
    0.06
     struggle
    0.06
     zeal
    0.06
    .Provider
    0.06
    Activations Density 0.003%

    No Known Activations