© 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. 102530
    Prev
    Next
    INDEX
    Explanations

    modelmachine learning models

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

    The neuron strongly responds to numeric tokens (digits, version numbers, years, and other numbers).

    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
    vario
    -0.77
    lotto
    -0.75
     robotic
    -0.74
    RG
    -0.74
    mapp
    -0.73
    廚
    -0.68
     RG
    -0.67
    inth
    -0.67
    asymp
    -0.67
    chter
    -0.66
    POSITIVE LOGITS
     inference
    0.98
    模型
    0.96
     모델
    0.94
     models
    0.92
     edge
    0.88
     ML
    0.87
     مدل
    0.87
     modelos
    0.85
    models
    0.84
     wake
    0.84
    Activations Density 0.040%

    No Known Activations