© 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. 10857
    Prev
    Next
    INDEX
    Explanations

    temporal sports events

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

    The neuron activates on numeric tokens—especially sports scores or statistical figures (including decimals).

    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
    acetic
    -0.94
    怎么回事
    -0.93
    individual
    -0.91
     erstaun
    -0.91
    NUC
    -0.91
     regalías
    -0.90
     months
    -0.89
    日程
    -0.88
    的眼神
    -0.86
    дио
    -0.86
    POSITIVE LOGITS
     tonight
    1.33
     yesterday
    1.31
     today
    1.25
     this
    1.21
     was
    1.20
     вчера
    1.14
     Friday
    1.08
     Saturday
    1.05
     était
    1.00
     Sunday
    0.97
    Activations Density 0.033%

    No Known Activations