© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13172
    Prev
    Next
    INDEX
    Explanations

    periods and commas

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron activates on numeric expressions and quantitative measurements (e.g. numbers, units, and decimal values).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     liefde
    -1.55
     gelar
    -1.53
     menikmati
    -1.49
     atlet
    -1.45
     दौर
    -1.43
     akrab
    -1.42
     kehadiran
    -1.38
     triko
    -1.37
    themselves
    -1.34
     TType
    -1.34
    POSITIVE LOGITS
     decided
    1.63
     (
    1.63
     then
    1.61
     didn
    1.47
     trochu
    1.44
     benutzt
    1.41
    嗰
    1.35
     couldn
    1.34
    那個
    1.33
    罷
    1.30
    Activations Density 0.109%

    No Known Activations