© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72631
    Prev
    Next
    INDEX
    Explanations

    animation and removal

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

    The neuron fires on programming‐style syntax (e.g. function calls, parentheses, braces, code keywords), effectively identifying 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_22/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
     rapidement
    -1.19
     ayrıca
    -1.12
     senare
    -1.01
     rychle
    -1.00
     Instantly
    -0.96
     entgegen
    -0.95
     immediatamente
    -0.94
     první
    -0.94
     sayesinde
    -0.94
    <bos>
    -0.94
    POSITIVE LOGITS
     another
    1.16
     and
    1.13
     after
    1.02
     it
    1.00
     will
    0.98
     continuing
    0.94
     continue
    0.94
     by
    0.93
     Geografia
    0.93
    פל
    0.93
    Activations Density 0.003%

    No Known Activations