© 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. 98320
    Prev
    Next
    INDEX
    Explanations

    large numbers and rounding

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

    The neuron fires on numeric literals (standalone numbers) in the text.

    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
     just
    -1.55
     if
    -1.52
     also
    -1.50
     any
    -1.49
     in
    -1.45
     one
    -1.41
     there
    -1.40
     even
    -1.38
     that
    -1.38
     by
    -1.36
    POSITIVE LOGITS
    0
    2.00
     horrid
    1.41
     wacky
    1.33
     sommige
    1.32
     flamboyant
    1.31
    upported
    1.29
     dismal
    1.28
     dreary
    1.27
     boister
    1.27
     revamped
    1.26
    Activations Density 0.027%

    No Known Activations