© 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. 30948
    Prev
    Next
    INDEX
    Explanations

    movie genres

    np_max-act · gemini-2.0-flash

    The neuron detects category labels describing film attributes and genres (e.g. “American,” “drama,” “comedy,” “English-language,” “black-and-white,” “upcoming”).

    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
    -na
    -0.08
     Burl
    -0.08
    _hidden
    -0.07
     Cup
    -0.07
     spread
    -0.06
    ınızda
    -0.06
    'ils
    -0.06
    _str
    -0.06
    .setHorizontalAlignment
    -0.06
     kwargs
    -0.06
    POSITIVE LOGITS
    RootElement
    0.07
    劳
    0.06
    Ltd
    0.06
     RemoteException
    0.06
     نظ
    0.06
     ministry
    0.06
     superiority
    0.06
    -grey
    0.06
    sher
    0.06
    SSL
    0.06
    Activations Density 0.006%

    No Known Activations