© 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. 22-GEMMASCOPE-RES-131K
    7. 35673
    Prev
    Next
    INDEX
    Explanations

    vis, type, input

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

    The neuron fires on all‐caps code tokens (acronyms or constant names, often with underscores).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     alcune
    -1.05
     certaines
    -1.05
    も
    -1.02
    があり
    -1.02
     acoper
    -1.02
     diferite
    -1.01
     alguns
    -1.01
     Dinas
    -0.98
     alcuni
    -0.96
    estim
    -0.94
    POSITIVE LOGITS
     QtCore
    1.20
     forState
    1.16
    ʽ
    1.08
    крови
    1.06
    szön
    1.06
    odeficiency
    1.04
     ""],
    1.03
    ゴリー
    1.03
    OrWhiteSpace
    1.02
    なのでしょう
    1.02
    Activations Density 0.006%

    No Known Activations