© 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. 28398
    Prev
    Next
    INDEX
    Explanations

    options or choice

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

    The neuron fires on license‐declaration phrases, signaling mentions of open‐source license names and related terms (e.g. “licensed under,” “BSD,” “GPL,” “MIT,” “dual licensed,” etc.).

    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
     meestal
    -0.86
    ્યા
    -0.85
     佐藤
    -0.82
     chodzi
    -0.82
     будь
    -0.81
     będę
    -0.81
     τὸν
    -0.79
    INGRED
    -0.79
    一定是
    -0.77
    [-\
    -0.77
    POSITIVE LOGITS
     alternative
    1.86
     options
    1.73
     alternatives
    1.73
     option
    1.70
    alternative
    1.43
     also
    1.38
     alternativas
    1.37
    Alternatively
    1.35
    options
    1.34
     alternatively
    1.34
    Activations Density 0.142%

    No Known Activations