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. 86273
    Prev
    Next
    INDEX
    Explanations

    There

    np_max-act · gemini-2.0-flash

    This neuron fires on structural markers and boundaries in the text—things like the end-of-text token, sentence-ending periods, and list or section intro words (e.g. “One,” “In,” “applications”) rather than on content words.

    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
    CppTypeDefinition
    -0.07
     wonders
    -0.06
     кілька
    -0.06
     Urs
    -0.06
    cores
    -0.06
     Apartment
    -0.06
     blasting
    -0.06
     CAUSED
    -0.06
     इसल
    -0.06
    .answer
    -0.06
    POSITIVE LOGITS
     teenagers
    0.08
    .isAdmin
    0.06
     revoked
    0.06
    -template
    0.06
     js
    0.06
    :x
    0.06
    交
    0.06
     năng
    0.06
    ße
    0.06
    obot
    0.06
    Activations Density 0.028%

    No Known Activations