© 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. 10-GEMMASCOPE-RES-131K
    7. 105826
    Prev
    Next
    INDEX
    Explanations

    superscript numbers or letters

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

    The neuron strongly activates on superscripted numeric markers (e.g. author affiliation or footnote numbers) in the document.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    焄
    -3.05
    s
    -2.52
    珖
    -2.39
     But
    -2.34
    <td>
    -2.33
     bolster
    -2.30
     However
    -2.28
     ALWAYS
    -2.25
     Since
    -2.23
    Donativos
    -2.23
    POSITIVE LOGITS
     ofrecen
    2.78
    
    2.67
    ሖ
    2.52
    ⢪
    2.44
    .
    2.31
    なども
    2.28
    ឵
    2.27
    垪
    2.27
     bedeutend
    2.25
     tratamientos
    2.20
    Activations Density 0.005%

    No Known Activations