© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128501
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is detect numeric tokens (digits and multi‐digit numbers) in mathematical expressions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    numbers followed by punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    に戻る
    -1.26
     pengusaha
    -1.18
     whoſe
    -1.14
     pemuda
    -1.13
    際に
    -1.10
    際には
    -1.09
     ieder
    -1.07
    に戻
    -1.06
    累了
    -1.06
     pueden
    -1.03
    POSITIVE LOGITS
     three
    1.16
     have
    1.08
     but
    1.07
    ってこと
    1.06
    〖
    1.05
    因为它
    1.03
     two
    1.02
    譜
    0.99
     lograr
    0.98
    その
    0.98
    Activations Density 0.023%

    No Known Activations