© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 0-GEMMASCOPE-TRANSCODER-16K
    7. 5661
    Prev
    Next
    INDEX
    Explanations

    text from a variety of sources that contain layout whitespace such as tabs, newlines and multiple spaces

    oai_token-act-pair · gemini-2.0-flash

    Technical/Legal text

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
     itſelf
    -1.58
     myſelf
    -1.54
      
    -1.45
     pleaſure
    -1.41
    ſelf
    -1.41
    ſelves
    -1.31
     Monfieur
    -1.30
     themſelves
    -1.28
     raiſ
    -1.27
     purpoſe
    -1.27
    POSITIVE LOGITS
     "
    0.76
     -
    0.73
     E
    0.73
     M
    0.73
    ,
    0.72
     G
    0.67
     ,
    0.67
     Ma
    0.66
     “
    0.66
     '
    0.66
    Activations Density 2.875%

    No Known Activations