© 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. 22-GEMMASCOPE-RES-131K
    7. 71017
    Prev
    Next
    INDEX
    Explanations

    after commas

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

    The main thing this neuron does is detect numeric tokens (multi‐digit numbers) in the text.

    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
     évent
    -1.34
    -
    -1.26
     différences
    -1.19
     nettoyer
    -1.18
    FRANC
    -1.16
    捂
    -1.15
     raconter
    -1.14
     botines
    -1.14
     compléter
    -1.13
     apporter
    -1.12
    POSITIVE LOGITS
     to
    2.11
     before
    1.87
     during
    1.79
     would
    1.69
     or
    1.65
     and
    1.61
     on
    1.55
     that
    1.52
     which
    1.52
     since
    1.50
    Activations Density 0.093%

    No Known Activations