© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 102042
    Prev
    Next
    INDEX
    Explanations

    measurement of time and speed

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

    The neuron fires on numeric tokens, especially measurements of time or counts (e.g., runtimes, durations, and other performance-related numbers).

    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
     mengal
    -1.39
     that
    -1.39
     kics
    -1.33
     escalera
    -1.29
     memastikan
    -1.23
     behov
    -1.21
    mereka
    -1.17
     suela
    -1.17
     дали
    -1.16
     laaj
    -1.16
    POSITIVE LOGITS
     and
    1.66
     fantasia
    1.25
     broderie
    1.24
     paradis
    1.19
     gamin
    1.17
     matel
    1.17
     aquarelle
    1.13
     commode
    1.12
     taupe
    1.11
     buttercream
    1.11
    Activations Density 0.072%

    No Known Activations