© 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. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 12622
    Prev
    Next
    INDEX
    Explanations

    general purpose words used in academic or technical writing such as "type of", "is a", and "proportional to".

    oai_token-act-pair · gemini-2.0-flash

    technical descriptions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.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
     Pakistani
    -1.26
     Arabic
    -1.26
     Iranian
    -1.25
     Iraqi
    -1.23
     Pakistan
    -1.18
     Saudi
    -1.17
     Arab
    -1.16
     Iran
    -1.14
     Arizona
    -1.13
     Arabs
    -1.13
    POSITIVE LOGITS
     resourceCulture
    0.54
    SharedDtor
    0.49
     RVA
    0.49
     Hepburn
    0.47
    
    0.46
    queryInterface
    0.45
     jsPsych
    0.44
     Fournier
    0.44
     eventdata
    0.44
     vermag
    0.44
    Activations Density 66.460%

    No Known Activations