© 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. 113237
    Prev
    Next
    INDEX
    Explanations

    news article reading times

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

    The neuron responds primarily to numeric tokens and quantitative expressions (e.g. dates, times, percentages, monetary amounts, “Min Read,” etc.).

    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
     cukru
    -2.02
     ragazza
    -1.97
    ׇ
    -1.96
     religieux
    -1.86
     katholischen
    -1.84
     sorella
    -1.83
     notte
    -1.79
     Erzb
    -1.77
     christlichen
    -1.76
    Válasz
    -1.74
    POSITIVE LOGITS
    ּוֹ
    2.23
    ֶּ
    1.93
    靭
    1.91
    וּ
    1.80
    ַּ
    1.79
    ִּ
    1.77
    靱
    1.73
     now
    1.73
     uitzicht
    1.69
    了个
    1.68
    Activations Density 0.146%

    No Known Activations