© 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. 68338
    Prev
    Next
    INDEX
    Explanations

    often, sometimes

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

    The neuron fires on numeric tokens (e.g. years, measurements, counts, or other numbers).

    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
     it
    -2.05
     two
    -1.95
     three
    -1.77
     one
    -1.47
     deportivas
    -1.46
     after
    -1.45
     people
    -1.41
     four
    -1.41
     started
    -1.39
     this
    -1.38
    POSITIVE LOGITS
    sometimes
    1.91
     parfois
    1.60
     sometimes
    1.59
    Sometimes
    1.58
    often
    1.52
     souvent
    1.48
     TÉCNICA
    1.48
     často
    1.47
     varier
    1.41
     daug
    1.35
    Activations Density 0.070%

    No Known Activations