© 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. 124104
    Prev
    Next
    INDEX
    Explanations

    code blocks and inline elements

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

    The neuron activates on strings that look like file or module path names in import/require/include statements.

    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
    
    -1.84
     '
    -1.81
    嘥
    -1.77
    alties
    -1.74
    跸
    -1.73
     あり
    -1.70
    が良い
    -1.66
    
    -1.66
    でご
    -1.64
     chargeur
    -1.64
    POSITIVE LOGITS
    7
    2.38
    ↵↵
    2.33
    3
    2.30
    2
    2.28
    …”
    2.27
    6
    2.11
     {
    2.09
    9
    2.08
    0
    2.06
    4
    2.02
    Activations Density 0.072%

    No Known Activations