© 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. 22-GEMMASCOPE-RES-131K
    7. 36604
    Prev
    Next
    INDEX
    Explanations

    joined after

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

    The neuron detects numeric time expressions—numbers that denote dates, durations, or ages (e.g. years, “five years ago,” “2001,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     will
    -1.27
    现在的
    -1.19
     Olimpia
    -1.13
     now
    -1.05
    常常
    -1.05
     Takt
    -1.03
     both
    -1.02
     murano
    -0.95
     sappi
    -0.93
     ducati
    -0.93
    POSITIVE LOGITS
     last
    2.47
     earlier
    1.90
     early
    1.67
     yesterday
    1.66
     after
    1.57
     recently
    1.53
     years
    1.50
     shortly
    1.49
    เมื่อ
    1.48
     hace
    1.41
    Activations Density 0.152%

    No Known Activations