© 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. 94178
    Prev
    Next
    INDEX
    Explanations

    capturing or recording screen

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

    The neuron fires on programming‐language code tokens and identifiers (e.g. “using,” “Bitmap,” include paths, method names, etc.), i.e. code snippets.

    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
     Provinz
    -0.72
    réhen
    -0.72
    Carn
    -0.69
     predstav
    -0.68
    íns
    -0.68
    Vz
    -0.68
     Rücks
    -0.66
     нев
    -0.66
     CIP
    -0.66
     DNA
    -0.65
    POSITIVE LOGITS
     capture
    1.58
    capture
    1.52
     capturing
    1.50
     recording
    1.34
    Capture
    1.29
     Capture
    1.26
     captures
    1.23
    capt
    1.16
     captur
    1.13
    recording
    1.11
    Activations Density 0.023%

    No Known Activations