© 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. 101937
    Prev
    Next
    INDEX
    Explanations

    Italian words and names

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

    The neuron activates on prominent content words—especially nouns and domain‐specific keywords that signal the main topic of each passage.

    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
     either
    -1.46
    cipy
    -1.26
     poate
    -1.14
     possibly
    -1.13
     dlou
    -1.12
     mãi
    -1.08
     září
    -1.06
     muertes
    -1.05
     oscuridad
    -1.05
     besonderes
    -1.04
    POSITIVE LOGITS
     romano
    1.34
     GRATIS
    1.23
     than
    1.21
    ...»
    1.17
     Novembre
    1.16
     diego
    1.15
     castel
    1.15
    2
    1.13
     fatale
    1.13
    ?»
    1.13
    Activations Density 0.432%

    No Known Activations