© 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. 70924
    Prev
    Next
    INDEX
    Explanations

    URLs and code references

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

    The neuron activates on numeric tokens (e.g. years, version numbers, percentages, and other digit sequences).

    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
     or
    -0.92
    apnews
    -0.91
    étro
    -0.90
     років
    -0.88
    nikon
    -0.88
    ecco
    -0.87
    avril
    -0.84
     penjualan
    -0.84
    出版年
    -0.84
     July
    -0.82
    POSITIVE LOGITS
    〬
    1.05
     kontin
    1.02
     themselves
    1.02
    anej
    1.00
    ಗ
    1.00
    herit
    1.00
    THERE
    0.99
    goes
    0.99
     rigu
    0.98
     compris
    0.96
    Activations Density 0.035%

    No Known Activations