© 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. 10-GEMMASCOPE-RES-131K
    7. 13577
    Prev
    Next
    INDEX
    Explanations

    so that

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

    The neuron strongly activates on numeric tokens—integers, decimals, and measurements—effectively detecting 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_10/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
     and
    -2.08
    在自己
    -1.58
    oooooo
    -1.55
    
    -1.52
     jf
    -1.50
     anun
    -1.49
    震惊
    -1.48
    being
    -1.48
     dahil
    -1.47
    doo
    -1.46
    POSITIVE LOGITS
    </h3>
    1.93
    殲
    1.70
     painstaking
    1.67
    </h4>
    1.66
    こんばん
    1.64
    歼
    1.59
     ingrained
    1.57
     ostensibly
    1.54
     dazzling
    1.53
    imetsu
    1.51
    Activations Density 0.017%

    No Known Activations