© 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. 24027
    Prev
    Next
    INDEX
    Explanations

    alteration or change

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

    The neuron identifies instances of numerical measurements or quantified changes (e.g. values, percentages, and decimal‐precision numbers) reported 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_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
    izing
    -1.30
    vetrain
    -1.27
     gioca
    -1.23
    現状
    -1.23
    ୌ
    -1.21
     svolge
    -1.20
     pommes
    -1.20
     villains
    -1.19
    ⋀
    -1.19
     racconta
    -1.16
    POSITIVE LOGITS
     each
    1.56
     we
    1.43
     depending
    1.37
    aconda
    1.34
     poszczegól
    1.31
     whereby
    1.30
     began
    1.29
     or
    1.27
     differently
    1.20
    liśmy
    1.20
    Activations Density 0.218%

    No Known Activations