© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41215
    Prev
    Next
    INDEX
    Explanations

    even as / though / while

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

    The neuron consistently activates on the adverb “even,” flagging its use as an emphatic modifier.

    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
    Also
    -1.63
     ukuran
    -1.59
    ”，
    -1.58
     who
    -1.57
     we
    -1.50
     Also
    -1.43
     setelan
    -1.39
     from
    -1.38
    and
    -1.35
     Çünkü
    -1.34
    POSITIVE LOGITS
     partici
    1.48
    </b>
    1.47
    promoting
    1.45
     beter
    1.42
    </strong>
    1.39
     extraordinaire
    1.38
     klink
    1.38
    吓得
    1.38
    ísmo
    1.36
    </i>
    1.34
    Activations Density 0.010%

    No Known Activations