© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91690
    Prev
    Next
    INDEX
    Explanations

    program code

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

    The neuron is triggered by programming‐style tokens—things like function or variable names, punctuation, and code keywords—i.e. snippets of source 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
     exuberant
    -0.82
    ̢
    -0.75
     interestingly
    -0.75
     adding
    -0.75
     strangely
    -0.75
     incorporate
    -0.74
    滨
    -0.74
     incorporating
    -0.73
    高雄
    -0.73
     dramatically
    -0.72
    POSITIVE LOGITS
     mouse
    1.04
     firmy
    0.99
    tess
    0.96
     presses
    0.94
     screenshot
    0.90
     SAYS
    0.88
     subprocess
    0.88
     OCR
    0.87
     Win
    0.86
    "]
    0.85
    Activations Density 0.016%

    No Known Activations