© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 92406
    Prev
    Next
    INDEX
    Explanations

    Entertainment

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on subword pieces that begin named entities or headings—i.e. capitalized tokens (proper‐noun prefixes like “Apatow,” “40‐Year,” “Ge” in “Geeks,” “Tele” in “Thymeleaf,” 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
    ouchers
    -0.07
     Manga
    -0.07
    وره
    -0.07
     gson
    -0.06
    ού
    -0.06
     Ко
    -0.06
    rypton
    -0.06
    agues
    -0.06
    .init
    -0.06
     glove
    -0.06
    POSITIVE LOGITS
    feel
    0.08
    (contract
    0.07
    iệp
    0.06
    Dice
    0.06
     Quantum
    0.06
    documents
    0.06
     dateTime
    0.06
     tik
    0.06
     tav
    0.06
    _php
    0.06
    Activations Density 0.059%

    No Known Activations