© 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. 67890
    Prev
    Next
    INDEX
    Explanations

    list items within code or data structures

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

    The neuron strongly activates on isolated single‐character tokens used as list or field labels—i.e. lone letters or digits that enumerate items.

    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
    heh
    -1.09
    向けの
    -1.02
    に入る
    -1.02
     the
    -1.00
    embang
    -0.96
    weren
    -0.95
     crucial
    -0.95
    solving
    -0.94
    quations
    -0.94
    cknowledge
    -0.93
    POSITIVE LOGITS
    wissenschaften
    1.09
    これは
    1.09
    тура
    0.95
     flera
    0.93
    reiben
    0.91
    的时候
    0.91
     rağmen
    0.91
    结尾
    0.91
    きましたが
    0.91
    США
    0.90
    Activations Density 0.063%

    No Known Activations