© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 102682
    Prev
    Next
    INDEX
    Explanations

    scientific notation and questions

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

    The neuron specifically fires on the “e” (and its associated sign and digits) when numbers are written in scientific/exponential notation.

    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
     so
    -0.85
    деля
    -0.82
     during
    -0.82
     often
    -0.80
     may
    -0.80
     or
    -0.80
     only
    -0.78
     early
    -0.78
    最も
    -0.77
     Often
    -0.76
    POSITIVE LOGITS
    wasn
    0.89
    dedicated
    0.88
    گي
    0.86
    𝑹
    0.83
    MainThread
    0.82
    ATHY
    0.82
     GRE
    0.81
     ADDED
    0.81
     風
    0.81
     vũ
    0.81
    Activations Density 0.004%

    No Known Activations