© 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 (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 13382
    Prev
    Next
    INDEX
    Explanations

    Introductions/overviews

    np_max-act · gemini-2.0-flash

    This neuron fires on meta‐text that frames or introduces instructional content—phrases like “This article will cover the tutorial…,” “have a look at it,” or “we have decided to provide you with a comprehensive summary.”

    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
     Griffin
    -0.08
     parc
    -0.07
     batter
    -0.07
     hobby
    -0.07
     prop
    -0.06
     ha
    -0.06
     disc
    -0.06
     travers
    -0.06
    اول
    -0.06
     Lac
    -0.06
    POSITIVE LOGITS
    .Now
    0.07
    ./
    0.07
    (lower
    0.06
    โรงเร
    0.06
    ใคร
    0.06
    DataContext
    0.06
     Львів
    0.06
    _ability
    0.06
    .master
    0.06
    Canonical
    0.06
    Activations Density 0.065%

    No Known Activations