© 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. 93091
    Prev
    Next
    INDEX
    Explanations

    years and numbers after words

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

    The neuron activates on numeric tokens—digits or numbers appearing in the text (e.g., years, counts, measurements).

    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
     With
    -1.90
     Since
    -1.61
     During
    -1.59
     has
    -1.58
     Before
    -1.52
     These
    -1.51
     does
    -1.49
     even
    -1.49
     Perhaps
    -1.46
     can
    -1.45
    POSITIVE LOGITS
     estadounid
    1.54
     legendar
    1.42
    providing
    1.41
     effici
    1.41
     éché
    1.40
     miniaturka
    1.37
     siè
    1.37
     téléph
    1.36
     misst
    1.36
    putting
    1.35
    Activations Density 0.064%

    No Known Activations