© 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. 22-GEMMASCOPE-RES-131K
    4. 73859
    Prev
    Next
    INDEX
    Explanations

    programming file extensions

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

    The neuron fires on standalone file‐name tokens—lines that consist of a filename (often in a comment or import/include) with its extension (e.g. “Layout.h”, “PrintModulePass.h”, “LUTheme.h”, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    
    -1.46
    なので
    -1.45
    ...........
    -1.29
    があり
    -1.26
     S
    -1.20
    ............
    -1.16
    
    -1.16
    
    -1.16
    武者
    -1.16
     "
    -1.16
    POSITIVE LOGITS
     belangrij
    1.62
    誰が
    1.61
    どうやって
    1.54
    ほら
    1.52
     plantillas
    1.46
    久し
    1.46
    の絵
    1.46
    angebote
    1.45
    為に
    1.44
     horarios
    1.42
    Activations Density 0.007%

    No Known Activations