© 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. 53979
    Prev
    Next
    INDEX
    Explanations

    Asteroid names, project design, Samsung, Q&A

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

    This neuron fires strongly on document section titles or metadata labels (e.g. headings like “Design” or codes like “SK”).

    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
     necessarie
    -1.00
    仍
    -0.98
     as
    -0.98
     during
    -0.97
     necessity
    -0.93
     čia
    -0.93
    neuri
    -0.91
    也是
    -0.91
    соеди
    -0.90
    كوم
    -0.88
    POSITIVE LOGITS
    oczes
    1.01
    aning
    1.00
    籲
    0.98
    aling
    0.97
     poursuite
    0.97
    的区别
    0.97
    麻煩
    0.95
     ort
    0.95
     mesures
    0.94
     orta
    0.93
    Activations Density 0.000%

    No Known Activations