© 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. 39575
    Prev
    Next
    INDEX
    Explanations

    research findings and analysis

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

    The neuron fires on terms that signal quantitative scientific reporting or measurement (e.g. “estimates,” “measured,” “reported,” “holds[the record],” “data,” “variables”).

    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
     生产
    -1.02
     calcu
    -1.00
    Dokument
    -0.97
    Biographical
    -0.94
    Gentile
    -0.91
    ܟ
    -0.91
    Причины
    -0.91
     divertir
    -0.90
    жник
    -0.89
    Guillermo
    -0.89
    POSITIVE LOGITS
    olm
    1.03
    eve
    1.03
     włosy
    1.01
     Blüten
    0.91
     of
    0.90
    governmental
    0.90
    Claim
    0.89
    Ð
    0.88
     mają
    0.88
    Applications
    0.88
    Activations Density 0.205%

    No Known Activations