© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 61447
    Prev
    Next
    INDEX
    Explanations

    reaction time, beat, scores

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

    The neuron strongly activates on numerical tokens and measurement-related terms (e.g. reaction, time, RT, percentages, scores), i.e. it identifies numerical data and quantitative 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_10/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
     蓝
    -1.20
     $\{$
    -1.19
     Heels
    -1.19
     ular
    -1.14
    נשים
    -1.13
     recommandations
    -1.12
     TLV
    -1.12
     langage
    -1.12
    surname
    -1.10
     chine
    -1.10
    POSITIVE LOGITS
     with
    1.56
     With
    1.26
     we
    1.24
     some
    1.20
     when
    1.19
     manages
    1.18
     provide
    1.15
     creato
    1.13
     crafted
    1.11
     doua
    1.09
    Activations Density 0.013%

    No Known Activations