© 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. 34-GEMMASCOPE-RES-131K
    7. 94214
    Prev
    Next
    INDEX
    Explanations

    URLs and social media links

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

    The neuron is tuned to recognize numeric tokens (sequences of digits) 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_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     устраи
    -0.89
     propriété
    -0.85
     followers
    -0.85
     Followers
    -0.81
    ió
    -0.80
     mats
    -0.77
     feed
    -0.76
    video
    -0.76
    𝔫
    -0.75
     Album
    -0.75
    POSITIVE LOGITS
    www
    1.64
     www
    1.32
    Monte
    0.82
    mounting
    0.81
    oglu
    0.81
     desee
    0.81
    ليج
    0.81
    cún
    0.80
    ثم
    0.77
    setInterval
    0.75
    Activations Density 0.023%

    No Known Activations