© 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. 26032
    Prev
    Next
    INDEX
    Explanations

    time and sequence

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

    The neuron consistently fires on numeric tokens—counts, percentages, measurements, reference numbers—i.e. any number in biomedical or scientific 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
    finally
    -1.23
    Finally
    -1.05
    мира
    -0.94
     enfin
    -0.94
    recently
    -0.86
    Neb
    -0.85
     finalmente
    -0.85
     наконец
    -0.83
     récemment
    -0.83
     endlich
    -0.82
    POSITIVE LOGITS
     then
    2.19
     thereafter
    1.89
     early
    1.87
     later
    1.63
     subsequent
    1.57
     Thereafter
    1.38
    その後
    1.34
     subsequently
    1.34
     immediately
    1.34
     затем
    1.32
    Activations Density 0.085%

    No Known Activations