© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 14529
    Prev
    Next
    INDEX
    Explanations

    ink jet printing

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

    The neuron fires on uncommon domain-specific jargon and proper names—particularly technical product or tool names (e.g. “watercolor,” “Eucalyptus tamint,” “tikz,” “inkjet,” artist names like Turner or Daniel Smith”).

    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
    饪
    -2.41
    ться
    -2.28
     }
    -2.16
    可以
    -2.13
    up
    -1.98
    </b>
    -1.94
    ’
    -1.94
    就是
    -1.93
    ०
    -1.88
     ---
    -1.84
    POSITIVE LOGITS
     菓子
    2.09
     Дне
    2.03
    蔔
    1.95
     墊
    1.93
    apatalk
    1.91
     rámci
    1.91
    acuation
    1.88
     beschäf
    1.87
     semelhantes
    1.84
    我现在
    1.84
    Activations Density 0.002%

    No Known Activations