© 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. 118887
    Prev
    Next
    INDEX
    Explanations

    MAINr = 5

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

    This neuron activates on unusually long sequences of digits (e.g. long numeric IDs) or similar long runs of punctuation/underscores.

    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
     Ebenen
    -0.82
    闡
    -0.78
    eler
    -0.77
     careful
    -0.76
    )։
    -0.75
     courteous
    -0.71
     Ortschaft
    -0.71
     obviously
    -0.70
    your
    -0.69
    )();
    -0.69
    POSITIVE LOGITS
     kuru
    0.99
    grandfather
    0.92
     didReceive
    0.88
     ۱۴
    0.86
    cardigan
    0.84
     mulai
    0.83
    lables
    0.82
    第一个
    0.82
     lancer
    0.82
    ᐠ
    0.82
    Activations Density 0.321%

    No Known Activations