© 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. 70662
    Prev
    Next
    INDEX
    Explanations

    average over all

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

    The neuron activates on numeric tokens (multi‐digit 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_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
    釆
    -1.14
     it
    -1.05
    upuncture
    -0.99
    stang
    -0.97
     vital
    -0.93
    Controllo
    -0.91
    nades
    -0.91
    ↵
    -0.90
     way
    -0.89
    ilogue
    -0.88
    POSITIVE LOGITS
    ślę
    1.29
     entire
    1.19
     jahre
    1.17
     calculées
    1.10
     meilleurs
    1.02
     etre
    1.00
     całej
    1.00
     hebben
    0.98
     horarios
    0.96
     vidéos
    0.95
    Activations Density 0.026%

    No Known Activations