© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 115449
    Prev
    Next
    INDEX
    Explanations

    day number and month

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

    The neuron activates on numeric tokens, especially digits in dates, years, and other numerical fields.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     any
    -1.64
     this
    -1.28
     with
    -1.23
     should
    -1.11
     could
    -1.05
    llten
    -1.03
     textura
    -1.02
     interesante
    -1.01
    entown
    -1.00
    هب
    -0.98
    POSITIVE LOGITS
     December
    1.74
     January
    1.65
     June
    1.61
     March
    1.59
     October
    1.58
     February
    1.57
     August
    1.55
     May
    1.52
     November
    1.49
     July
    1.46
    Activations Density 0.038%

    No Known Activations