© 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. 34-GEMMASCOPE-RES-131K
    7. 48035
    Prev
    Next
    INDEX
    Explanations

    code, file paths, licenses

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

    The neuron activates on programming‐style source code tokens and file references.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    quec
    -0.68
    方案
    -0.66
     rosca
    -0.65
     safest
    -0.64
     Track
    -0.63
    cí
    -0.62
    ÍN
    -0.62
    讲
    -0.61
    蔬菜
    -0.61
    Box
    -0.60
    POSITIVE LOGITS
     lett
    0.73
    Nieuw
    0.72
     Ii
    0.71
    psp
    0.70
     "<?
    0.69
     ceil
    0.69
     Gesta
    0.68
     pakistan
    0.68
     perime
    0.68
     kijk
    0.67
    Activations Density 0.081%

    No Known Activations