© 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. 71756
    Prev
    Next
    INDEX
    Explanations

    HTML form tags

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

    The neuron activates on numeric tokens—i.e. it’s looking for number literals wherever they appear.

    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
    Indien
    -0.91
     mangá
    -0.89
    豈
    -0.84
    PECT
    -0.82
     bloco
    -0.81
    Weiß
    -0.81
    ibilità
    -0.77
     следующем
    -0.76
     Sünde
    -0.76
     サイズ
    -0.76
    POSITIVE LOGITS
    gradable
    0.94
    enek
    0.89
    liath
    0.87
    DDDD
    0.86
     Kalender
    0.86
     erstat
    0.84
    lty
    0.84
     kvinder
    0.84
    anka
    0.82
     withRouter
    0.82
    Activations Density 0.041%

    No Known Activations