© 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. 10-GEMMASCOPE-RES-131K
    7. 60536
    Prev
    Next
    INDEX
    Explanations

    M/C. or s/n

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

    The neuron fires strongly on isolated single-letter tokens and short letter-slash abbreviations (e.g. “C”, “w”, “N/A”, “M/s”), i.e. technical shorthand or code-like fragments.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     However
    -2.67
     helped
    -2.53
     allows
    -2.52
     kept
    -2.45
     reduces
    -2.44
     prevents
    -2.44
     epitom
    -2.42
     promotes
    -2.38
     markedly
    -2.33
     went
    -2.30
    POSITIVE LOGITS
     of
    3.11
     توضی
    2.44
     amd
    2.09
     REALLY
    2.02
     soooo
    1.98
    匆
    1.98
     insanely
    1.98
     unn
    1.97
     flashy
    1.95
     revamped
    1.93
    Activations Density 0.042%

    No Known Activations