© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99899
    Prev
    Next
    INDEX
    Explanations

    ratio per unit

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

    The neuron activates on numeric expressions—digits, number words, and monetary amounts—used for counts, percentages, and financial figures.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     ogrom
    -0.87
     Übungen
    -0.86
     plusieurs
    -0.82
    nearly
    -0.82
    vasti
    -0.82
     الطريق
    -0.81
     چند
    -0.81
     hizo
    -0.81
    rentices
    -0.81
     tématu
    -0.80
    POSITIVE LOGITS
     one
    1.63
     every
    1.37
     each
    1.31
     One
    1.27
     dollar
    1.10
     ONE
    1.09
     каж
    1.02
    one
    0.99
     ஒரு
    0.98
     hour
    0.97
    Activations Density 0.050%

    No Known Activations