© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14262
    Prev
    Next
    INDEX
    Explanations

    then later events

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

    The neuron fires on numeric tokens—especially years or other date numbers.

    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
    diese
    -1.32
    leicht
    -1.27
     salonu
    -1.23
    leten
    -1.21
     kommende
    -1.20
    welche
    -1.16
    …»
    -1.15
    zyj
    -1.13
    können
    -1.13
    meine
    -1.11
    POSITIVE LOGITS
     further
    1.47
     etiam
    1.23
     then
    1.20
     later
    1.19
     joined
    1.16
     official
    1.13
     wider
    1.12
     formal
    1.11
     expanded
    1.10
     expand
    1.09
    Activations Density 0.087%

    No Known Activations