© 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. 68356
    Prev
    Next
    INDEX
    Explanations

    event completion or transition

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

    the neuron detects words that mark significant time references or temporal events (e.g., war, season, years, deadline, election, weather, time).

    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
     folo
    -1.21
     prek
    -1.16
     studen
    -1.14
     vla
    -1.10
    футболка
    -1.09
     ordina
    -1.08
     ショーツ
    -1.05
     sada
    -1.05
    անի
    -1.05
     髪型
    -1.04
    POSITIVE LOGITS
     came
    1.29
     I
    1.18
     M
    1.08
     time
    1.07
     out
    1.00
     S
    0.98
     Did
    0.97
     D
    0.94
     insist
    0.93
    してる
    0.93
    Activations Density 0.030%

    No Known Activations