© 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. 22-GEMMASCOPE-RES-131K
    4. 70608
    Prev
    Next
    INDEX
    Explanations

    transition to current information

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

    The neuron fires on discourse-transition adverbs—especially the sentence-initial “Now” that shifts or continues the narrative.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     vẫn
    -1.21
     still
    -1.13
     знают
    -1.07
    still
    -1.07
    captures
    -1.06
     kellett
    -1.05
    相変わらず
    -1.05
     هنوز
    -1.04
     actuellement
    -1.03
     ήταν
    -1.03
    POSITIVE LOGITS
     comes
    1.20
     роки
    1.19
     wants
    1.14
    さらに
    1.12
     with
    1.06
     eius
    1.02
    ՝
    1.00
     explicou
    1.00
     seeks
    0.95
    ,
    0.94
    Activations Density 0.028%

    No Known Activations