© 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. 28420
    Prev
    Next
    INDEX
    Explanations

    questions and specific topics

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

    The neuron strongly activates on tokens that are purely numeric (multi‐digit number strings).

    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
     of
    -1.98
     your
    -1.04
    нциклопе
    -1.02
     utaf
    -1.00
    ↵↵
    -0.97
     habari
    -0.94
     identify
    -0.93
    もちろん
    -0.92
     kazi
    -0.92
     Regione
    -0.90
    POSITIVE LOGITS
     nového
    1.38
     veranderen
    1.33
     ܕ
    1.20
    少々
    1.18
     verwijderen
    1.17
     dobré
    1.17
     nových
    1.16
     čtyř
    1.14
    を見ると
    1.13
     intervalo
    1.09
    Activations Density 0.001%

    No Known Activations