© 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. 104666
    Prev
    Next
    INDEX
    Explanations

    temporal sequence and timing

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

    The neuron specializes in picking out schedule-related date and time mentions (e.g. calendar dates, meeting times, months, 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
     сейчас
    -1.13
     diferentes
    -1.05
     maintenant
    -1.05
     distintas
    -1.05
     often
    -1.03
     teilweise
    -0.98
    now
    -0.97
     différentes
    -0.96
     now
    -0.95
     distintos
    -0.95
    POSITIVE LOGITS
     thereafter
    1.88
     then
    1.83
     afterwards
    1.70
     Thereafter
    1.55
     next
    1.54
     потім
    1.41
     afterward
    1.39
    然后
    1.35
     Then
    1.32
     THEN
    1.30
    Activations Density 0.031%

    No Known Activations