© 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. 22-GEMMASCOPE-RES-131K
    7. 14139
    Prev
    Next
    INDEX
    Explanations

    medical, scientific, technical terms

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

    scientific or technical terms within academic or medical research contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron specializes in spotting quantitative or numerical information—particularly measurements, counts, and other data-rich tokens—in technical and scientific 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_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
     และ
    -1.03
    に行って
    -0.96
     And
    -0.95
    !,
    -0.88
    正常
    -0.87
     tho
    -0.85
    glLoadIdentity
    -0.85
    そして
    -0.84
    并将
    -0.83
     な
    -0.83
    POSITIVE LOGITS
     ہے۔
    1.29
     $_=
    1.00
     atk
    0.95
     among
    0.94
     boken
    0.93
     romantis
    0.92
     ہیں۔
    0.91
    りました
    0.91
     mosaico
    0.90
    っていました
    0.89
    Activations Density 0.165%

    No Known Activations