© 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. 34-GEMMASCOPE-RES-131K
    7. 76035
    Prev
    Next
    INDEX
    Explanations

    create/set up technical terms

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

    The neuron fires on imperative verbs in code comments (e.g. “Get,” “Set,” “Stop,” “Serialize,” “Submit”) that signal step‐by‐step instructions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    supposed
    -1.17
    こんなに
    -1.14
    Garantie
    -1.09
    salsa
    -1.05
    numerus
    -1.02
    かっこ
    -1.01
     affitto
    -1.01
    Menge
    -1.00
    Temperatura
    -1.00
    Programma
    -1.00
    POSITIVE LOGITS
     our
    1.52
    érienne
    1.20
    (':
    1.17
    λ
    1.17
    
    1.16
    انجليز
    1.13
     '
    1.13
    n
    1.12
    de
    1.10
    Our
    1.09
    Activations Density 0.035%

    No Known Activations