© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 23605
    Prev
    Next
    INDEX
    Explanations

    highest mountains and peaks

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

    The main thing this neuron does is detect numeric tokens (e.g. sequences of digits representing measurements, counts, dates, etc.).

    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
    ἃ
    -0.96
    ുറ
    -0.90
     seashore
    -0.88
     quartered
    -0.87
     оружи
    -0.86
     Shoreline
    -0.85
     ধ
    -0.85
    Litteratur
    -0.83
    研究生
    -0.82
    lectic
    -0.81
    POSITIVE LOGITS
     summit
    2.17
     peak
    1.71
     tallest
    1.63
     peaks
    1.60
     height
    1.52
     summits
    1.52
     Peak
    1.50
     elevation
    1.41
    highest
    1.39
     highest
    1.35
    Activations Density 0.060%

    No Known Activations