© 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. 39771
    Prev
    Next
    INDEX
    Explanations

    copyright notice, conditions, disclaimer

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

    The neuron selectively activates on numeral tokens, i.e. standalone multi-digit numbers in the text.

    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
    来週
    -2.91
     and
    -2.88
    明日は
    -2.50
    マジで
    -2.38
    ところに
    -2.36
    ツイッター
    -2.36
    さっそく
    -2.34
    おそ
    -2.31
    で見る
    -2.31
    久し
    -2.30
    POSITIVE LOGITS
    もある
    2.14
    ⟞
    2.00
     þe
    1.95
    が
    1.93
    も
    1.87
    がある
    1.84
    を
    1.84
    鼇
    1.82
     lyd
    1.82
    <bos>
    1.77
    Activations Density 0.001%

    No Known Activations