© 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. 128519
    Prev
    Next
    INDEX
    Explanations

    cle

    np_max-act · gemini-2.0-flash

    This neuron responds to numeric and measurement tokens (especially digits and decimal-point values describing attributes like age, height, weight, cup size, etc.).

    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
    Term
    -0.07
     LESS
    -0.07
    기의
    -0.06
    ディア
    -0.06
    _vert
    -0.06
    adata
    -0.06
    ;};↵
    -0.06
     clan
    -0.06
    pson
    -0.06
    _fm
    -0.06
    POSITIVE LOGITS
     kart
    0.07
     NotImplementedException
    0.07
     oblivious
    0.07
     Bark
    0.06
     scal
    0.06
    TagName
    0.06
    .alignment
    0.06
    、今
    0.06
    getContent
    0.06
    vetica
    0.06
    Activations Density 0.026%

    No Known Activations