© 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. 105440
    Prev
    Next
    INDEX
    Explanations

    experiencing or having states

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

    The neuron strongly activates on verbs in their “-ing” (gerund/participle) form, especially common state‐ or experience‐denoting ones like “being,” “having,” and “experiencing.”

    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
     our
    -1.95
     other
    -1.83
     those
    -1.78
     that
    -1.73
     they
    -1.60
     what
    -1.59
     just
    -1.53
     first
    -1.47
    ;
    -1.46
    yoruz
    -1.45
    POSITIVE LOGITS
    こいつ
    1.59
    possibly
    1.52
    自己是
    1.45
     to
    1.41
     Bemerkungen
    1.40
    section
    1.38
     künf
    1.37
     gegenwär
    1.34
     incongru
    1.33
    being
    1.30
    Activations Density 0.039%

    No Known Activations