© 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. 22-GEMMASCOPE-RES-131K
    7. 39881
    Prev
    Next
    INDEX
    Explanations

    Russian geography and photography

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

    The neuron detects informal discourse markers and interjections (like “Hahaha,” “Okay,” “Anyway,” etc.) that signal conversational asides or transitions.

    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
    ঋ
    -1.14
    ére
    -1.06
     نیروهای
    -1.02
    था
    -1.01
     SEVENTH
    -1.00
    före
    -0.97
    aquí
    -0.96
    逃げ
    -0.96
     گردد
    -0.94
     LYON
    -0.94
    POSITIVE LOGITS
    inerary
    1.23
     nordrhein
    1.21
     география
    1.17
    aphthalene
    1.17
     знают
    1.16
     {$\
    1.15
     которы
    1.15
     фотография
    1.14
    .‏
    1.14
     lopp
    1.13
    Activations Density 0.004%

    No Known Activations