© 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. 47220
    Prev
    Next
    INDEX
    Explanations

    time ranges

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

    The neuron strongly activates on standalone numeric tokens (e.g. years, counts, IDs, or other multi‐digit numbers).

    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
     Additionally
    -1.15
     "[
    -1.02
     approximately
    -1.02
     decreased
    -0.98
     amongst
    -0.98
     ladies
    -0.97
    casted
    -0.97
    “[
    -0.96
     due
    -0.95
     “[
    -0.93
    POSITIVE LOGITS
    这里是
    1.02
     meantime
    0.94
     sejumlah
    0.94
     România
    0.90
     aici
    0.89
     depois
    0.89
     нынеш
    0.88
     Надо
    0.86
     coluna
    0.86
    nahm
    0.86
    Activations Density 0.043%

    No Known Activations