© 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. 10-GEMMASCOPE-RES-131K
    7. 75140
    Prev
    Next
    INDEX
    Explanations

    mechanisms underlying phenomena

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

    This neuron triggers on words that signal foundational explanation or knowledge—especially “underlying” (and related notions like “know”) indicating a search for basic mechanisms or understanding.

    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
     protéger
    -1.47
    probably
    -1.40
    )),
    -1.39
     cruda
    -1.38
     trente
    -1.36
    鏽
    -1.35
    沵
    -1.34
    GIVEREF
    -1.33
     réinitial
    -1.32
    鸸
    -1.32
    POSITIVE LOGITS
     where
    2.11
     with
    1.97
     how
    1.48
     even
    1.48
    Carreira
    1.35
     on
    1.33
     all
    1.27
    Према
    1.26
    イン
    1.26
     potrzeby
    1.24
    Activations Density 0.071%

    No Known Activations