© 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. 34-GEMMASCOPE-RES-131K
    7. 75408
    Prev
    Next
    INDEX
    Explanations

    logical, mathematical, and formal language structures

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

    This neuron activates on LaTeX‐style math tokens—especially numerals, subscripts (e.g. “_”, digits), and backslashes used in 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_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
     Thursday
    -1.05
     krás
    -0.98
    liebe
    -0.95
     randonnée
    -0.93
     картофе
    -0.92
    ğim
    -0.91
    𝐆
    -0.91
     творо
    -0.91
    ⊻
    -0.90
     insiste
    -0.90
    POSITIVE LOGITS
     syntactic
    0.99
     AttributeSet
    0.91
     Roscoe
    0.91
    Bool
    0.91
    ugais
    0.89
    ¬
    0.88
     ¬
    0.87
    新鮮
    0.86
     ornamentation
    0.86
     refuted
    0.85
    Activations Density 0.006%

    No Known Activations