© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 13684
    Prev
    Next
    INDEX
    Explanations

    XML and code attributes

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

    The neuron strongly activates on numeric literal tokens (i.e. any standalone numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    indentation or whitespace at the beginning of lines in code or structured text files.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    structural formatting and syntax tokens from code/markup snippets, especially indentation/whitespace and special symbols.

    oai_token-act-pair · gpt-5Triggered 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
     to
    -1.05
    *}[
    -1.00
     that
    -0.98
     TAMBIÉN
    -0.94
     schließlich
    -0.94
     until
    -0.93
     botón
    -0.88
     if
    -0.85
     there
    -0.85
     both
    -0.83
    POSITIVE LOGITS
     maniere
    1.12
     ongles
    1.07
     ngan
    1.04
    onej
    1.01
     élas
    0.96
     orale
    0.94
     nhan
    0.93
    vily
    0.93
     VIET
    0.93
     sentier
    0.93
    Activations Density 0.001%

    No Known Activations