© 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. 41452
    Prev
    Next
    INDEX
    Explanations

    straight year or season

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

    The neuron activates on words and phrases that convey quantity, degree, or numerical context (e.g. “many,” “more,” “straight,” “on pace,” etc.).

    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
     Minist
    -1.16
     peeking
    -1.08
     faltar
    -1.04
    Osm
    -1.02
    Ital
    -1.02
     почти
    -1.01
     solcher
    -0.99
     Institu
    -0.94
    ¹,
    -0.94
    Coro
    -0.93
    POSITIVE LOGITS
    ereço
    1.30
    ього
    1.24
     negocios
    1.15
     exposição
    1.13
    pography
    1.12
     inspección
    1.10
    cts
    1.08
     parfaite
    1.08
    rab
    1.07
    delar
    1.07
    Activations Density 0.198%

    No Known Activations