© 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. 13695
    Prev
    Next
    INDEX
    Explanations

    actions connected by and

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

    The neuron primarily picks out numeric tokens (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_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
     siempre
    -1.10
    jahres
    -1.07
    Réponse
    -1.05
     sprawling
    -1.04
     Jednak
    -1.02
    繋
    -1.02
    often
    -1.02
    BIBLIOGRAPHY
    -1.01
    迎来
    -1.00
     ""){
    -1.00
    POSITIVE LOGITS
     trying
    1.74
     đang
    1.66
     doing
    1.63
     están
    1.49
     estão
    1.47
     attempting
    1.35
     estamos
    1.30
     getting
    1.24
     aren
    1.21
     preparing
    1.21
    Activations Density 0.129%

    No Known Activations