© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14499
    Prev
    Next
    INDEX
    Explanations

    Gmail, iced tea, Lucknow

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

    The neuron strongly activates on standalone nouns that are names or labels—especially proper names, brands, and technical acronyms.

    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
    </td>
    -3.02
    肫
    -2.91
     is
    -2.83
    ”.
    -2.69
    辚
    -2.67
    0
    -2.64
    l
    -2.61
    «
    -2.61
    盦
    -2.59
    ⸨
    -2.52
    POSITIVE LOGITS
    躋
    2.52
     замет
    2.48
    一旁
    2.39
    少爷
    2.30
    ?"
    2.30
     serves
    2.25
     vsak
    2.25
    旁边的
    2.22
    lenül
    2.20
     hangs
    2.19
    Activations Density 0.005%

    No Known Activations