© 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. 68943
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and operations

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

    This neuron activates on TeX/LaTeX command tokens (backslash‐initiated math macros) within mathematical expressions.

    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
    🎦
    -1.55
     consé
    -1.53
    そして
    -1.52
    ですが
    -1.48
    voer
    -1.46
    なぜ
    -1.41
    ﾚ
    -1.41
    dagogik
    -1.41
    また
    -1.40
    も
    -1.40
    POSITIVE LOGITS
    為に
    2.02
     aanzien
    1.48
     vormen
    1.48
     overeen
    1.45
    ために
    1.45
    様な
    1.45
    ようです
    1.42
    方がいい
    1.41
    に入れる
    1.38
     wodurch
    1.37
    Activations Density 0.007%

    No Known Activations