© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 74629
    Prev
    Next
    INDEX
    Explanations

    a bit of

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

    The neuron detects adverbs or modifiers meaning “a small amount” (e.g. French peu, Spanish poco, English bit, Russian немного).

    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
     neuen
    -2.55
    鱻
    -2.41
    してくれた
    -2.06
     verifiable
    -2.05
     Instead
    -2.03
     biggest
    -2.02
    んびり
    -2.00
    屢
    -1.94
    父の
    -1.94
     schnellen
    -1.94
    POSITIVE LOGITS
    «
    2.63
     GOtt
    2.41
     «
    2.28
    「
    2.19
    儅
    2.16
    klärung
    2.08
     animo
    2.03
    <bos>
    2.03
     Monfieur
    2.02
     teneb
    2.00
    Activations Density 0.002%

    No Known Activations