© 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. 10-GEMMASCOPE-RES-131K
    4. 42261
    Prev
    Next
    INDEX
    Explanations

    further + verb indicating additional action

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

    This neuron responds to longer, multi-syllabic technical or formal terms—especially participial or nominal words—that typically appear as functional or transitional vocabulary in patent-style documents.

    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
     always
    -1.35
     while
    -1.30
     this
    -1.24
    ]}{
    -1.19
     definitely
    -1.18
     in
    -1.07
    mentable
    -1.07
     educativas
    -1.05
     just
    -1.05
    while
    -1.05
    POSITIVE LOGITS
     jeweiligen
    1.45
     navíc
    1.30
    -
    1.22
     luxuri
    1.13
    runners
    1.12
     étui
    1.10
     freakin
    1.10
    ující
    1.10
    dert
    1.08
     inoltre
    1.08
    Activations Density 0.064%

    No Known Activations