© 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. 39688
    Prev
    Next
    INDEX
    Explanations

    transmitting data or information

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

    The neuron is specialized to detect hyphenated compound descriptors (e.g. “network-based,” “in-vivo”) in technical or patent-style text.

    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
     
    -1.15
    ем
    -1.01
     「
    -0.98
     faktiskt
    -0.98
     chinois
    -0.98
     composants
    -0.97
     superstars
    -0.96
     are
    -0.93
    chuhe
    -0.93
    employee
    -0.92
    POSITIVE LOGITS
     who
    1.34
     by
    1.26
     entsprechenden
    1.11
     interested
    1.09
     other
    1.09
     соответству
    1.06
     prosegu
    1.05
     appropriate
    1.03
     运动
    1.02
    適切
    1.02
    Activations Density 0.063%

    No Known Activations