© 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. 102334
    Prev
    Next
    INDEX
    Explanations

    diagrams, lists, charts, maps, networks

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

    The neuron responds to mentions of visual‐diagram constructs—words naming charts, maps, diagrams, and related plotting or mapping terms.

    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
     edin
    -1.35
     telles
    -1.34
     GOING
    -1.28
     effectuer
    -1.27
    cisive
    -1.25
     veren
    -1.24
    accurate
    -1.23
    coherent
    -1.21
     donner
    -1.21
    should
    -1.21
    POSITIVE LOGITS
     about
    2.47
     new
    1.58
     particularly
    1.46
     especially
    1.45
     mainly
    1.41
     around
    1.38
     not
    1.36
     이야
    1.36
     different
    1.34
     would
    1.34
    Activations Density 0.052%

    No Known Activations