© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 29158
    Prev
    Next
    INDEX
    Explanations

    parts of a system

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

    The neuron strongly responds to abstract, multi-syllable nouns—particularly those ending in “-ion”—that denote processes, measurements, or other technical/academic concepts.

    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
     divertente
    -1.25
     prachtig
    -1.22
    как
    -1.20
     trasparente
    -1.17
     pulsante
    -1.16
     Fridge
    -1.16
     vlees
    -1.14
     дії
    -1.12
     invito
    -1.12
    zestaw
    -1.12
    POSITIVE LOGITS
     the
    1.58
    ),
    1.45
     após
    1.23
     помню
    1.21
    ’,
    1.20
     remember
    1.18
     meinem
    1.18
     dopo
    1.15
    Nieuwe
    1.12
    いいですね
    1.10
    Activations Density 1.069%

    No Known Activations