© 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. 68042
    Prev
    Next
    INDEX
    Explanations

    numbers, punctuation, and measurements

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

    The neuron activates on tokens composed of or containing numeric digits (i.e. multi-digit numbers or numeric strings).

    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
     dejo
    -0.90
     univerz
    -0.87
     balon
    -0.86
    });
    -0.86
    yenne
    -0.85
    Volkswagen
    -0.85
    StandardCharsets
    -0.85
     ſo
    -0.85
    Koordinaten
    -0.82
    ound
    -0.82
    POSITIVE LOGITS
    明確
    0.98
     to
    0.91
    明日
    0.89
    𝔭
    0.88
    They
    0.85
    ek
    0.85
    杩
    0.85
     To
    0.84
     
    0.83
    There
    0.81
    Activations Density 0.004%

    No Known Activations