© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 10204
    Prev
    Next
    INDEX
    Explanations

    STEM and related fields

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

    The neuron detects occurrences of the acronym “STEM” (often in contexts like STEM Education or related educational/technical mentions).

    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
    乆
    -2.16
     Using
    -2.14
    इटम
    -1.95
    痦
    -1.89
    够了
    -1.85
    单击
    -1.76
    }
    -1.75
     Trying
    -1.73
     Having
    -1.73
     Considering
    -1.73
    POSITIVE LOGITS
    蟶
    2.14
    setBold
    2.00
    と思い
    1.98
     définiti
    1.96
    ⃨
    1.86
     Inſ
    1.86
    鹛
    1.85
     grafico
    1.84
     of
    1.83
    o
    1.83
    Activations Density 0.003%

    No Known Activations