© 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. 440
    Prev
    Next
    INDEX
    Explanations

    mathematical formulas

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

    The neuron fires strongly on numeric tokens and mathematical symbols in formulas—that is, it identifies numbers and related equation components.

    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
    ᾿
    -1.05
     ſi
    -1.02
     behandelt
    -1.02
     komplette
    -1.01
    bedien
    -0.99
    strator
    -0.99
    modellen
    -0.98
    HttpStatusCode
    -0.97
     sigo
    -0.96
     Nederlands
    -0.96
    POSITIVE LOGITS
     for
    1.15
     with
    1.08
     into
    1.00
     erstmals
    0.99
     at
    0.99
    餚
    0.98
    ibold
    0.95
     capable
    0.93
     their
    0.91
     expressed
    0.90
    Activations Density 0.005%

    No Known Activations