© 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. 34-GEMMASCOPE-RES-131K
    7. 93779
    Prev
    Next
    INDEX
    Explanations

    numbers and codes

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

    This neuron activates on numeric tokens—that is, it spotlights sequences of digits (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_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
    efel
    -0.87
     Intellectual
    -0.85
    anam
    -0.84
     indie
    -0.83
     даже
    -0.82
    olta
    -0.81
    嗓
    -0.80
     Finanzierung
    -0.79
    ዘ
    -0.78
    missible
    -0.78
    POSITIVE LOGITS
    句話
    0.93
     its
    0.89
    はその
    0.85
     their
    0.85
     ihrer
    0.79
    醜
    0.79
     hau
    0.76
    practi
    0.75
     OY
    0.74
     スクール
    0.74
    Activations Density 0.015%

    No Known Activations