© 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. 22-GEMMASCOPE-RES-131K
    4. 13790
    Prev
    Next
    INDEX
    Explanations

    SANTA BARBARA

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

    The neuron selectively fires on standalone numeric tokens—especially version‐style numbers in package/dependency metadata.

    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
     cuando
    -1.36
     when
    -1.35
     would
    -1.18
     it
    -1.15
     there
    -1.15
     fueron
    -1.06
     starts
    -1.04
     begins
    -1.04
     began
    -1.03
    noční
    -1.02
    POSITIVE LOGITS
     maniere
    1.38
     appels
    1.34
     vanille
    1.34
     vœ
    1.27
     frambo
    1.27
     réinitial
    1.27
     horloge
    1.26
     honneur
    1.21
     applau
    1.20
     reager
    1.20
    Activations Density 0.002%

    No Known Activations