© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26890
    Prev
    Next
    INDEX
    Explanations

    excess, surplus, leftover, unused

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

    The neuron strongly activates on numeric tokens or quantity-related figures (e.g. digits, decimals, large number substrings).

    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
    ↹
    -1.15
     apes
    -1.11
     verv
    -1.11
     Sust
    -1.09
    cester
    -1.08
    upport
    -1.07
     Bibl
    -1.07
     SUPPORT
    -1.05
     EFF
    -1.04
     Twain
    -1.04
    POSITIVE LOGITS
     after
    1.74
     leftover
    1.55
     excess
    1.54
     that
    1.41
     unused
    1.37
     when
    1.33
     sometimes
    1.21
     surplus
    1.16
     last
    1.13
    Unused
    1.13
    Activations Density 0.032%

    No Known Activations