© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 16775
    Prev
    Next
    INDEX
    Explanations

    closing html tags

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

    The main thing this neuron does is find numeric tokens (numbers) in the text.

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     such
    -1.05
     tramonto
    -0.96
     for
    -0.96
     first
    -0.95
    żdż
    -0.94
    챕
    -0.93
    ceptre
    -0.92
     silicona
    -0.92
     one
    -0.90
    ……。
    -0.89
    POSITIVE LOGITS
    </strong>
    3.03
    </b>
    2.69
    </h2>
    2.36
    </em>
    2.33
    </h4>
    2.27
    </u>
    2.11
    </i>
    2.08
    </h3>
    1.91
    </h1>
    1.84
    </h5>
    1.63
    Activations Density 0.110%

    No Known Activations