© 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. 10815
    Prev
    Next
    INDEX
    Explanations

    code punctuation and citations

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

    The neuron activates on digit‐only tokens (numeric literals) in code.

    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
     lecz
    -1.11
     in
    -1.10
     лъ
    -1.04
     while
    -1.02
     lakini
    -1.02
     years
    -1.01
     médic
    -0.97
    净化
    -0.95
     antecedent
    -0.94
     oreilles
    -0.93
    POSITIVE LOGITS
     приготовить
    1.16
     Produkten
    1.09
     apropiado
    1.05
    virano
    1.04
    ことになる
    1.03
    ];
    1.02
    ועל
    1.02
     обеспечить
    1.01
     beispielsweise
    1.00
    ulée
    1.00
    Activations Density 0.001%

    No Known Activations