© 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. 104618
    Prev
    Next
    INDEX
    Explanations

    months and seasons

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

    This neuron is detecting relatively uncommon (i.e. “rare”) tokens—words with high inverse-document-frequency.

    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
     tary
    -1.01
     sali
    -1.00
     сначала
    -0.99
     compuestos
    -0.98
     quando
    -0.98
    클
    -0.97
     comprenant
    -0.97
     lorsqu
    -0.93
     complementos
    -0.93
     lej
    -0.92
    POSITIVE LOGITS
     January
    2.06
     September
    2.03
     July
    2.02
     December
    2.02
     November
    1.98
     winter
    1.95
     October
    1.95
     February
    1.92
     during
    1.91
     summer
    1.90
    Activations Density 0.157%

    No Known Activations