© 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
    2. Gemma-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 10108
    Prev
    Next
    INDEX
    Explanations

    mostly proper nouns that are geographical locations, especially locations in Europe and Malaysia

    oai_token-act-pair · gemini-2.0-flash

    geographical locations

    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
    ViewImports
    -0.59
     continúas
    -0.56
    :✨
    -0.50
    ografija
    -0.48
     صوتيه
    -0.47
    プーン
    -0.47
     gunner
    -0.45
    Demografie
    -0.44
    uxxxx
    -0.44
     متعلقه
    -0.43
    POSITIVE LOGITS
    ChildScrollView
    0.71
     Plain
    0.59
    itespace
    0.57
    Rujuakan
    0.56
    Kanpo
    0.56
    脚注の使い方
    0.54
    +:+
    0.54
     National
    0.52
    laka
    0.52
    PickerController
    0.51
    Activations Density 1.210%

    No Known Activations