© 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. 98494
    Prev
    Next
    INDEX
    Explanations

    separators and conjunctions

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

    The main thing this neuron does is detect numeric tokens (e.g. measurements, counts, and other number values).

    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
    だね
    -1.20
     gdyż
    -1.16
    たびに
    -1.14
     wiedzieć
    -1.13
    感じです
    -1.12
     
    -1.11
    )।
    -1.11
     belangrij
    -1.10
     These
    -1.09
     следующих
    -1.08
    POSITIVE LOGITS
    OPPO
    1.20
     ͡°
    1.15
     psychiat
    1.15
     سایر
    1.14
     benzin
    1.13
     荣耀
    1.13
     velours
    1.10
     plafon
    1.10
     壁纸
    1.09
     Mâ
    1.09
    Activations Density 0.001%

    No Known Activations