© 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. 53861
    Prev
    Next
    INDEX
    Explanations

    initialisms and abbreviations

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

    The neuron strongly activates on brief alphanumeric “code‐like” tokens—short sequences of letters and digits used as identifiers, acronyms, file names, docket numbers, or phone-number fragments—rather than ordinary prose.

    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
     ogóle
    -0.93
    alte
    -0.79
    种类
    -0.77
     بالإضافة
    -0.75
    醍
    -0.75
    Gover
    -0.75
     each
    -0.74
     Teach
    -0.73
    叁
    -0.72
     principals
    -0.71
    POSITIVE LOGITS
    ｉ
    0.94
     unwise
    0.88
    obviously
    0.88
    tiness
    0.85
    sw
    0.84
    ards
    0.84
     tepi
    0.84
    ending
    0.83
     Jawahar
    0.82
     allerlei
    0.82
    Activations Density 0.029%

    No Known Activations