© 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. 93227
    Prev
    Next
    INDEX
    Explanations

    numbers and structural markers

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

    The neuron strongly activates on standalone digits (or digit sequences) when they appear as part of alphanumeric identifiers or codes, e.g. version or model numbers.

    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
     aneh
    -0.98
    Morfo
    -0.92
     camisetas
    -0.91
    梧
    -0.88
     lombok
    -0.88
     ladr
    -0.87
    tanleria
    -0.86
    ㈬
    -0.85
    Tür
    -0.84
    sprzed
    -0.84
    POSITIVE LOGITS
     непосредственно
    0.95
     же
    0.94
    ？』
    0.93
    ressed
    0.90
    More
    0.88
    ig
    0.86
    im
    0.86
    もう少し
    0.86
    その他の
    0.85
     against
    0.84
    Activations Density 0.005%

    No Known Activations