© 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. 60760
    Prev
    Next
    INDEX
    Explanations

    templating syntax like @, {, {{, {%

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

    The neuron strongly activates on code‐specific punctuation and templating markers (e.g. “@”, “{{”, “(”) and other syntax tokens, effectively identifying snippets of source or template code.

    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
     hvilke
    -0.84
     小型
    -0.81
     WITHOUT
    -0.81
     toit
    -0.77
     dij
    -0.76
    ppchen
    -0.76
     berbagai
    -0.76
    èle
    -0.76
    ไหน
    -0.76
     surn
    -0.75
    POSITIVE LOGITS
    '><
    0.89
    "><
    0.85
     bảng
    0.84
     forgiving
    0.83
    gmt
    0.82
    龑
    0.81
    -"
    0.80
    #__
    0.79
     snippetHide
    0.78
    ሀ
    0.78
    Activations Density 0.000%

    No Known Activations