© 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. 2-GEMMASCOPE-TRANSCODER-16K
    7. 11435
    Prev
    Next
    INDEX
    Explanations

    words appearing in code and certain Portuguese words, or possibly just the word "do" and certain suffixes

    oai_token-act-pair · gemini-2.0-flash

    do

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_2/width_16k/average_l0_49
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.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
     do
    -2.38
     Do
    -1.73
    do
    -1.71
    Do
    -1.71
     DO
    -1.44
    DO
    -1.27
     does
    -1.19
     did
    -1.13
     doing
    -0.98
    <bos>
    -0.93
    POSITIVE LOGITS
     ivelany
    0.72
    XmlRootElement
    0.64
     HasFactory
    0.64
     Económica
    0.60
    WebVitals
    0.60
     Drafting
    0.59
     lccn
    0.59
     Reforms
    0.58
     Disqus
    0.57
     Perennial
    0.57
    Activations Density 0.981%

    No Known Activations