© 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. 75038
    Prev
    Next
    INDEX
    Explanations

    shell commands and code snippets

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

    The neuron strongly responds to the names of shell‐ or UNIX‐style command‐line utilities (e.g. awk, cut, grep, sed, du, wc, etc.) when they appear 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
     might
    -1.45
     When
    -1.20
     equipado
    -1.20
     silencioso
    -1.19
     protegido
    -1.19
     elástico
    -1.15
    mö
    -1.13
     this
    -1.12
     alarg
    -1.12
     should
    -1.11
    POSITIVE LOGITS
    amazing
    1.16
     публикации
    1.13
     nejen
    1.13
    LAST
    1.13
    entown
    1.09
    vatel
    1.08
     начинает
    1.07
     maxime
    1.06
     APPROPRI
    1.06
     angep
    1.05
    Activations Density 0.023%

    No Known Activations