© 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. 10-GEMMASCOPE-RES-131K
    7. 19210
    Prev
    Next
    INDEX
    Explanations

    time or schedule duration

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

    The neuron fires on numeric tokens and words that introduce or quantify time (like “timeline,” “timeframe,” “timescale,” and the associated digits).

    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
    Etimología
    -1.76
    leggio
    -1.61
    Unlike
    -1.60
    囖
    -1.52
    Despite
    -1.52
    越来越多
    -1.49
    bernador
    -1.49
    apparition
    -1.44
    酝
    -1.44
     There
    -1.42
    POSITIVE LOGITS
    an
    2.13
    of
    1.80
    憷
    1.79
    迓
    1.74
    x
    1.71
     oppure
    1.63
    liath
    1.60
    臯
    1.59
    暃
    1.59
    le
    1.59
    Activations Density 0.014%

    No Known Activations