© 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. 34-GEMMASCOPE-RES-131K
    7. 99252
    Prev
    Next
    INDEX
    Explanations

    bounded

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

    This neuron selectively responds to mathematical notation—particularly LaTeX-style formula tokens (symbols, numeric fractions, variable names, dollar‐delimited math) and equation keywords.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     merciless
    -0.77
     bezeich
    -0.74
     초
    -0.72
    HEAD
    -0.70
    市内
    -0.69
    HPLC
    -0.69
     Expenses
    -0.68
     chosen
    -0.68
    supers
    -0.67
     imposto
    -0.67
    POSITIVE LOGITS
     bounded
    2.00
    bounded
    1.73
    Bounded
    1.48
     unbounded
    1.16
     Bound
    1.15
     bound
    1.05
    BOUND
    0.98
     Cauchy
    0.93
    Bound
    0.91
     bounds
    0.87
    Activations Density 0.035%

    No Known Activations