© 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. 22-GEMMASCOPE-RES-131K
    7. 32947
    Prev
    Next
    INDEX
    Explanations

    Save big

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

    The neuron is specialized to pick out standalone numeric tokens (numbers) in the text.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     condiciones
    -0.93
     there
    -0.90
     exceptions
    -0.89
     Gebäude
    -0.88
     escase
    -0.86
     dual
    -0.85
    ATEUR
    -0.84
     daarna
    -0.84
     resultado
    -0.84
     receiving
    -0.84
    POSITIVE LOGITS
    Assuming
    1.09
    Whether
    1.02
     '',
    0.98
    主要是
    0.98
    Mostly
    0.97
    Mainly
    0.97
    Normally
    0.97
    Either
    0.96
    ibr
    0.96
    Usually
    0.95
    Activations Density 0.000%

    No Known Activations