© 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. 39298
    Prev
    Next
    INDEX
    Explanations

    analysis and computation

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

    The neuron strongly activates on specialized technical or scientific terms (e.g. comparative adverbs in scientific descriptions, biochemical suffixes, code identifiers), i.e. jargon or domain‐specific vocabulary.

    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
     would
    -1.13
    ‘
    -1.06
     While
    -0.97
    While
    -0.93
    â
    -0.92
    「
    -0.89
     lightly
    -0.87
    “
    -0.86
     subtle
    -0.86
     â
    -0.85
    POSITIVE LOGITS
     són
    1.05
     salonu
    0.96
    などが
    0.96
     BEAUT
    0.96
     klasa
    0.95
     SOCIAL
    0.94
     zobac
    0.93
    навли
    0.93
     naše
    0.92
     dát
    0.92
    Activations Density 0.019%

    No Known Activations