© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 0-GEMMASCOPE-TRANSCODER-16K
    7. 14115
    Prev
    Next
    INDEX
    Explanations

    various forms of the "ing" suffix when talking about a process and references to something increasing

    oai_token-act-pair · gemini-2.0-flash

    ding

    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
    Negative Logits
    ding
    -0.66
    ة
    -0.66
    .
    -0.62
    ing
    -0.56
     libras
    -0.55
    avras
    -0.55
    <unused61>
    -0.51
    o
    -0.51
    ted
    -0.50
     dioses
    -0.50
    POSITIVE LOGITS
    存于互联网档案馆
    1.05
     Roskov
    1.01
    abestanden
    0.98
    InstrumentedTest
    0.94
    Brainz
    0.92
    Personensuche
    0.87
     MaterialApp
    0.86
     OFDb
    0.84
     defaultstate
    0.82
    featureID
    0.82
    Activations Density 2.828%

    No Known Activations