© 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. 9166
    Prev
    Next
    INDEX
    Explanations

    research proposals

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens naming academic‐research elements or structural sections (e.g. “research,” “literature,” “methodology,” “study,” “analysis,” “data,” “paper,” “dissertation,” “proposal”).

    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
     ################################
    -0.07
     stmt
    -0.07
     evet
    -0.07
     parody
    -0.07
    Country
    -0.07
    .email
    -0.06
     BOOK
    -0.06
    .logo
    -0.06
     ring
    -0.06
     manufactures
    -0.06
    POSITIVE LOGITS
     نیم
    0.07
    達
    0.07
    逃
    0.06
    LowerCase
    0.06
     spectators
    0.06
    ynthia
    0.06
    κρα
    0.06
     控
    0.06
     обы
    0.06
    أم
    0.06
    Activations Density 0.024%

    No Known Activations