© 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. 26267
    Prev
    Next
    INDEX
    Explanations

    numbers and months

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

    The neuron fires strongly on numeric date components (days, months, and years) in timeline or event descriptions.

    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
     each
    -1.23
     new
    -1.20
     as
    -1.15
     this
    -1.12
     for
    -1.06
     during
    -1.05
     if
    -1.02
     including
    -1.01
     ideas
    -1.00
     every
    -0.99
    POSITIVE LOGITS
     炒め
    1.16
    mochila
    1.15
    icali
    1.13
     cumplido
    1.13
     uiteindelijk
    1.13
     rencontrer
    1.12
     finalement
    1.06
    received
    1.05
     recibirá
    1.04
     medzi
    1.03
    Activations Density 0.101%

    No Known Activations