© 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. 123582
    Prev
    Next
    INDEX
    Explanations

    water or chemical contexts

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

    references to water and water-related terms, including chemical compounds, solutions, and water-based substances.

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

    The neuron strongly activates on occurrences of the word “water” (including when it appears in compounds like “water‐soluble”).

    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
    in
    -1.77
    ’
    -1.72
    it
    -1.59
    ing
    -1.54
     saying
    -1.48
    9
    -1.47
    1
    -1.47
    ‘
    -1.46
    ss
    -1.45
     itself
    -1.45
    POSITIVE LOGITS
     různ
    1.55
     zlep
    1.45
    黼
    1.43
     нужна
    1.40
     únicas
    1.35
    Inggris
    1.34
    بيل
    1.34
     příjem
    1.30
     мальчик
    1.30
    Comté
    1.28
    Activations Density 0.057%

    No Known Activations