© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 60598
    Prev
    Next
    INDEX
    Explanations

    merging items, data, or code

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

    The neuron activates on occurrences of the root “merge” (and its variants like “merges”) in code, comments, and identifiers.

    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
    ↵↵
    -3.08
    en
    -3.02
    訁
    -2.78
    光芒
    -2.70
    ꔹ
    -2.67
    癭
    -2.61
    鱟
    -2.58
     voor
    -2.58
     we
    -2.55
     same
    -2.48
    POSITIVE LOGITS
    寕
    3.08
    她也
    2.48
    ますが
    2.41
    
    2.38
    蹀
    2.27
     從
    2.25
    先は
    2.22
    次回は
    2.20
    ijnlijk
    2.19
    みんなで
    2.14
    Activations Density 0.008%

    No Known Activations