© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 48005
    Prev
    Next
    INDEX
    Explanations

    act of inserting

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

    The neuron primarily activates on the word “insert” (and its variants such as “insertion”).

    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
    牘
    -1.02
     سقف
    -1.02
    眎
    -0.99
    ઁ
    -0.98
     deniz
    -0.98
    Història
    -0.98
    pkins
    -0.96
    ?!”
    -0.96
     urges
    -0.95
     tula
    -0.95
    POSITIVE LOGITS
     into
    1.78
     Insert
    1.54
     插入
    1.27
     inserting
    1.27
     insert
    1.21
    insert
    1.16
    InsertCommand
    1.13
     inser
    1.13
    OrUpdate
    1.13
     inserted
    1.11
    Activations Density 0.019%

    No Known Activations