© 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. 34-GEMMASCOPE-RES-131K
    7. 73339
    Prev
    Next
    INDEX
    Explanations

    combining things to create results

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

    This neuron detects summative “it all adds up to…”–style connective phrases that signal “X all adds up to Y.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Likewise
    -1.15
     JpaRepository
    -1.02
     Likewise
    -0.99
    volves
    -0.98
     utilizan
    -0.98
    same
    -0.97
    Similarly
    -0.95
     suivants
    -0.94
     involves
    -0.94
    chene
    -0.93
    POSITIVE LOGITS
     combine
    2.14
     combination
    1.73
    Combine
    1.70
     recipe
    1.63
     all
    1.60
     Combine
    1.58
     combining
    1.56
     altogether
    1.48
    combine
    1.45
     ergibt
    1.41
    Activations Density 0.052%

    No Known Activations