© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68643
    Prev
    Next
    INDEX
    Explanations

    "so" followed by a descriptor

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

    The neuron fires on strong intensifier words or degree modifiers (e.g. “so,” “excited,” “overwhelmingly”) that amplify or emphasize a statement.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    Ако
    -1.73
     spese
    -1.71
     aziende
    -1.71
    さぁ
    -1.67
     qualsiasi
    -1.63
     mattina
    -1.55
     réun
    -1.55
     creazione
    -1.55
    no
    -1.49
    にん
    -1.46
    POSITIVE LOGITS
     and
    2.33
     or
    2.20
     also
    2.00
     или
    1.82
    </h3>
    1.63
     monitores
    1.48
     exemples
    1.44
     atau
    1.41
     juist
    1.34
     evidently
    1.32
    Activations Density 0.016%

    No Known Activations