© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72254
    Prev
    Next
    INDEX
    Explanations

    adverbs describing state or completion

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

    The neuron strongly activates on hyphens or dashes—particularly when they appear in hyphenated words or numeric ranges.

    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
    さてさて
    -1.27
    それにしても
    -1.26
    どうやら
    -1.25
    というわけで
    -1.23
    itisation
    -1.17
    や
    -1.16
    izacji
    -1.16
    をご覧ください
    -1.15
     recevrez
    -1.13
    Ār
    -1.13
    POSITIVE LOGITS
    為に
    1.11
    世的
    0.98
    ,
    0.96
     ebenfalls
    0.93
     Similarly
    0.93
    H
    0.93
     字母
    0.91
     functioning
    0.91
     别
    0.91
    ริม
    0.90
    Activations Density 0.076%

    No Known Activations