© 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. 68189
    Prev
    Next
    INDEX
    Explanations

    scientific or technical terms

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

    The neuron is looking for numeric tokens (instances of numbers) in the text.

    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
     öde
    -1.20
    skap
    -1.20
    guidelines
    -1.13
    𐀀
    -1.05
    München
    -1.00
    zhu
    -0.98
     CardView
    -0.97
    ╙
    -0.97
     PopupWindow
    -0.97
     $@
    -0.96
    POSITIVE LOGITS
     instead
    1.00
     as
    0.98
     but
    0.94
    つまり
    0.91
     other
    0.90
     besides
    0.85
     before
    0.84
    内容は
    0.84
     thinking
    0.83
     among
    0.82
    Activations Density 0.002%

    No Known Activations