© 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. 15695
    Prev
    Next
    INDEX
    Explanations

    project summary

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

    The neuron fires on uppercase code identifiers that begin with “PROJECT_,” i.e. tokens matching the PROJECT_… pattern.

    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
     you
    -2.02
     モード
    -1.91
    蹋
    -1.90
     เพราะ
    -1.88
     Потому
    -1.85
     Mitgliedern
    -1.84
     Некоторые
    -1.82
     ваш
    -1.81
    <td>
    -1.79
     Многие
    -1.77
    POSITIVE LOGITS
    ?”
    2.52
    ሷ
    2.31
    ement
    2.08
     ensamb
    2.06
    
    2.03
     materiais
    1.99
    䉴
    1.97
    珧
    1.96
    彞
    1.94
    ?</
    1.93
    Activations Density 0.003%

    No Known Activations