© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14632
    Prev
    Next
    INDEX
    Explanations

    numerical calculations

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

    The neuron responds to standalone numeric tokens (multi‐digit 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
     so
    -1.27
    詆
    -1.16
    klima
    -1.11
     klima
    -1.05
     seksual
    -1.02
     vitale
    -1.02
     brune
    -1.02
    konstru
    -1.02
    moderne
    -1.02
     zog
    -1.02
    POSITIVE LOGITS
     planejamento
    1.05
     indivíduo
    1.02
     goofy
    0.99
     astounding
    0.96
     gigantic
    0.96
     anúncio
    0.96
    都快
    0.95
    可以用
    0.94
     insane
    0.94
     trabajando
    0.93
    Activations Density 0.028%

    No Known Activations