© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 19164
    Prev
    Next
    INDEX
    Explanations

    International words, dates, punctuation

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

    The neuron strongly activates on the digit “2” whenever it occurs as part of an ISO-style timestamp (e.g. in “2015-02-27T22:28:00.5839218”).

    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
     if
    -1.95
     If
    -1.83
     types
    -1.79
     kinds
    -1.74
     are
    -1.71
     for
    -1.67
     wichtigsten
    -1.65
    ными
    -1.65
     Sumatera
    -1.59
    inée
    -1.57
    POSITIVE LOGITS
    .
    1.73
     parfü
    1.57
    »,
    1.44
     tejido
    1.41
     моей
    1.40
    うつ
    1.40
    坏了
    1.38
    礼物
    1.38
    悲伤
    1.37
     женщины
    1.34
    Activations Density 0.030%

    No Known Activations