© 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. 22-GEMMASCOPE-RES-131K
    7. 84053
    Prev
    Next
    INDEX
    Explanations

    invention relate

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

    This neuron strongly activates on standalone numeric tokens (digits) used as IDs, indices, or codes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     抱
    -0.99
     landes
    -0.98
    に
    -0.96
    ъ
    -0.92
    ské
    -0.92
     Hidup
    -0.92
    arschijnlijk
    -0.91
    oogle
    -0.90
    zellan
    -0.90
    tà
    -0.90
    POSITIVE LOGITS
     れる
    0.94
    ניין
    0.93
     hopeful
    0.92
    買う
    0.92
     atractivos
    0.91
     others
    0.90
     many
    0.87
     mân
    0.87
     PHILOSOPHY
    0.86
     mourir
    0.85
    Activations Density 0.000%

    No Known Activations