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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 54097
    Prev
    Next
    INDEX
    Explanations

    months of the year

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

    The neuron fires on date-related tokens, particularly month names and adjacent numeric date components.

    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
     both
    -1.05
     will
    -1.02
     see
    -1.01
     twice
    -1.01
     has
    -0.95
     have
    -0.93
     below
    -0.93
     regarding
    -0.93
     methods
    -0.92
     sostanza
    -0.92
    POSITIVE LOGITS
     nouvelles
    1.07
     seguras
    1.07
    Released
    1.07
    rozum
    1.06
    ᾖ
    1.05
    λλην
    1.03
     ürün
    1.03
     espera
    1.02
     appétit
    1.02
     Expected
    0.98
    Activations Density 0.059%

    No Known Activations