© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 6463
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to legal proceedings and illegal activity, including trade and transportation of goods.

    oai_token-act-pair · gemini-2.0-flash

    Illegal transportation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pemasaran
    -0.49
     propriétaires
    -0.47
     ketahui
    -0.47
     bolsillos
    -0.46
     sèche
    -0.46
     aislados
    -0.45
     attentes
    -0.44
     ejecut
    -0.44
     auszu
    -0.44
     démon
    -0.43
    POSITIVE LOGITS
     cigars
    0.73
     vuitton
    0.72
     firearms
    0.70
     ""],
    0.69
     تانيه
    0.68
     disambiguazione
    0.67
     cigarettes
    0.67
     NUK
    0.65
     valuables
    0.64
     gunpowder
    0.64
    Activations Density 5.976%

    No Known Activations