© 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. 130268
    Prev
    Next
    INDEX
    Explanations

    titles

    np_max-act · gemini-2.0-flash

    The neuron fires on genre and subgenre labels—especially hyphenated or multiword genre names (e.g. “Planet Sci-Fi,” “Artificial Intelligence Sci-Fi,” “Automaton Steampunk”).

    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
     lower
    -0.07
    VML
    -0.07
    aying
    -0.07
     školy
    -0.07
    rst
    -0.06
     knife
    -0.06
     glue
    -0.06
    Maximum
    -0.06
     button
    -0.06
     refuge
    -0.06
    POSITIVE LOGITS
     Subscriber
    0.07
     Ire
    0.06
     subsidi
    0.06
    _excerpt
    0.06
     Mort
    0.06
    consulta
    0.06
     didReceiveMemoryWarning
    0.06
     Vick
    0.06
     acknow
    0.06
    λι
    0.06
    Activations Density 0.029%

    No Known Activations