© 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. 583
    Prev
    Next
    INDEX
    Explanations

    package structure

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

    This neuron activates on numeric tokens, especially those forming dates, times, and other multi-digit numbers.

    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
     you
    -1.21
     cilind
    -1.10
     alip
    -1.09
     Gottesdienst
    -1.08
     misi
    -1.06
     COLORS
    -1.05
     tuta
    -1.04
     diamant
    -1.03
     certainly
    -1.03
     surely
    -1.01
    POSITIVE LOGITS
     miej
    1.30
    וּ
    1.18
    Procedimento
    1.16
    tersebut
    1.15
    Informazioni
    1.14
    Beratung
    1.11
    nunca
    1.10
    camin
    1.06
    Dimensões
    1.06
     lekko
    1.05
    Activations Density 0.005%

    No Known Activations