© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72778
    Prev
    Next
    INDEX
    Explanations

    timing before events

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

    The neuron consistently fires on numeric tokens (including integers and decimals), i.e. it detects numbers and quantities 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
     spion
    -0.97
    erende
    -0.96
     continuare
    -0.91
    Progress
    -0.89
    progress
    -0.88
     sako
    -0.86
    urezza
    -0.86
    wagon
    -0.84
     Göttingen
    -0.84
    monium
    -0.83
    POSITIVE LOGITS
     beforehand
    3.11
     prior
    2.55
     vorher
    2.48
     заранее
    2.45
     preceding
    2.38
     earlier
    2.38
     trước
    2.16
    事前に
    2.06
     before
    2.03
    提前
    1.98
    Activations Density 0.105%

    No Known Activations