© 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. 35099
    Prev
    Next
    INDEX
    Explanations

    specific concepts or items

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

    The neuron strongly activates on numeric tokens (e.g. any standalone numbers or digit sequences such as page numbers, counts, or citation indices).

    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
     Meksiko
    -1.12
     kompeti
    -1.06
     histó
    -1.02
     dalam
    -0.99
     konfer
    -0.98
     pouvaient
    -0.97
     asfal
    -0.97
     ucran
    -0.96
     meren
    -0.96
    鹟
    -0.96
    POSITIVE LOGITS
     its
    0.97
     overzicht
    0.94
     our
    0.94
     sisält
    0.92
     kierunku
    0.83
     käyt
    0.83
     elämä
    0.83
    𝚜
    0.82
     revolving
    0.82
    ियां
    0.82
    Activations Density 0.054%

    No Known Activations