© 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. 107334
    Prev
    Next
    INDEX
    Explanations

    numbers and counts

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

    The neuron is primarily detecting numerical data points—counts, statistics, or quantities—often those that follow verbs or prepositions.

    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
     several
    -1.26
     dozens
    -1.16
     twenties
    -1.12
    十
    -1.07
     half
    -1.07
     two
    -1.02
     tens
    -1.01
    Several
    -0.99
     >
    -0.98
     three
    -0.98
    POSITIVE LOGITS
    exactly
    1.29
     insgesamt
    1.19
    total
    1.07
     fewer
    1.06
     huiles
    1.03
    ģ
    1.02
     총
    1.00
    người
    0.99
     části
    0.98
     seventy
    0.97
    Activations Density 0.491%

    No Known Activations