© 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. 10-GEMMASCOPE-RES-131K
    7. 41460
    Prev
    Next
    INDEX
    Explanations

    specific layer names or identifiers

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

    The neuron is detecting programming tokens that define or reference neural network layers (e.g. “layer,” layer indices, and related parameter fields).

    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
     bookId
    -1.58
    即是
    -1.56
     predecess
    -1.48
    蓊
    -1.48
     которой
    -1.46
     \\
    -1.46
     would
    -1.43
    “.
    -1.42
    लिए
    -1.40
     getM
    -1.39
    POSITIVE LOGITS
     harten
    1.63
     to
    1.57
     fünf
    1.49
     také
    1.47
     neun
    1.45
     pertama
    1.45
    唯美
    1.45
     you
    1.42
     też
    1.41
     разделе
    1.41
    Activations Density 0.003%

    No Known Activations