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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89375
    Prev
    Next
    INDEX
    Explanations

    igration

    np_max-act · gemini-2.0-flash

    This neuron activates on identifiers in code—especially mixed-case class or method names (e.g. Schemamigration, migrate, Animation, SharedConsumer).

    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
    Friends
    -0.07
    aying
    -0.07
    works
    -0.06
    となる
    -0.06
     Albania
    -0.06
    iology
    -0.06
    Embed
    -0.06
     blood
    -0.06
    SETTING
    -0.06
    Focused
    -0.06
    POSITIVE LOGITS
     supportive
    0.08
    iants
    0.07
    <x
    0.06
     income
    0.06
     možnosti
    0.06
    .modify
    0.06
     conclus
    0.06
    ,state
    0.06
    $j
    0.06
    _animation
    0.06
    Activations Density 0.067%

    No Known Activations