© 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. 10-GEMMASCOPE-RES-131K
    7. 101870
    Prev
    Next
    INDEX
    Explanations

    green

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

    The neuron is highly responsive to mentions of color names—especially the word “green.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    </b>
    -2.19
     in
    -2.05
    </u>
    -2.05
     “
    -2.02
    jenige
    -1.95
     to
    -1.92
     was
    -1.87
    럿
    -1.84
    d
    -1.77
     out
    -1.76
    POSITIVE LOGITS
     gantung
    2.69
     ideen
    2.61
     剧照
    2.41
    橚
    2.38
     normaux
    2.31
    獁
    2.31
    禇
    2.22
     mengucap
    2.22
    They
    2.20
    0
    2.19
    Activations Density 0.005%

    No Known Activations