© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99527
    Prev
    Next
    INDEX
    Explanations

    backward

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

    The neuron fires on numeric expressions—quantities, measurements, dates, monetary amounts, percentages, 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
    Достоинства
    -0.83
    ńskiego
    -0.82
    🍨
    -0.77
     Fei
    -0.77
     sabi
    -0.77
    អ
    -0.76
     خبرگزاری
    -0.75
    ταν
    -0.75
    ivably
    -0.73
    Appellee
    -0.73
    POSITIVE LOGITS
     backward
    1.18
    Backward
    1.04
     captive
    1.01
    Heidelberg
    0.97
    backward
    0.95
    standalone
    0.94
     Setup
    0.94
     Backward
    0.93
     segment
    0.91
    靭
    0.90
    Activations Density 0.014%

    No Known Activations