© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113595
    Prev
    Next
    INDEX
    Explanations

    end of year/month/date

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

    The neuron strongly responds to numeric expressions of dates or time—especially year numbers and words like “end” when they occur in temporal phrases.

    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
     inteligen
    -1.60
    戋
    -1.58
    鹋
    -1.54
     intelek
    -1.52
     zí
    -1.52
    ܖ
    -1.49
    </sup>
    -1.48
     pomá
    -1.46
    ⧠
    -1.46
     negatif
    -1.45
    POSITIVE LOGITS
    2
    1.96
    ↵
    1.70
    in
    1.62
    However
    1.49
     there
    1.49
    7
    1.47
    6
    1.47
    0
    1.40
    Despite
    1.39
    Here
    1.34
    Activations Density 0.011%

    No Known Activations