© 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. 12720
    Prev
    Next
    INDEX
    Explanations

    code structure identifiers

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

    The neuron fires on uppercase framework/decorator identifiers—e.g. Java/Kotlin annotation names (Entity, PrimaryKey, ColumnInfo) or similar imported decorator class names.

    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
    красивые
    -1.44
     unangemessene
    -1.36
    ութ
    -1.33
    thodoxy
    -1.21
     entendimiento
    -1.14
     asalto
    -1.13
     знаешь
    -1.12
    悧
    -1.11
    ammlung
    -1.10
     apropi
    -1.09
    POSITIVE LOGITS
     and
    1.38
    nós
    1.30
     decât
    1.20
    konos
    1.18
    ongiorno
    1.18
     nicht
    1.16
    colat
    1.13
    łach
    1.13
    只有一个
    1.10
    likely
    1.09
    Activations Density 0.040%

    No Known Activations