© 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. 14463
    Prev
    Next
    INDEX
    Explanations

    FAQ and questions

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

    The neuron triggers on structural markers of a FAQ or numbered question list, i.e. tokens that label and punctuate individual questions.

    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
    YYCURSOR
    -1.37
     an
    -1.37
      “
    -1.35
     ---
    -1.35
    šana
    -1.34
    belajaran
    -1.34
     also
    -1.33
    欵
    -1.33
     також
    -1.30
    Odpovědět
    -1.30
    POSITIVE LOGITS
     incredi
    1.40
     egna
    1.39
    各样的
    1.26
     dotyczą
    1.26
    fusions
    1.24
     incess
    1.22
     incluse
    1.20
    0
    1.18
    cologist
    1.18
     huit
    1.17
    Activations Density 0.008%

    No Known Activations