© 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. 75351
    Prev
    Next
    INDEX
    Explanations

    months in dates

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

    The neuron detects date expressions, especially month names and their associated numeric day or year tokens.

    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
    tivis
    -1.00
     cáncer
    -0.92
     Públicas
    -0.90
    ά
    -0.90
    ołnier
    -0.90
    jutkan
    -0.89
     μαζί
    -0.88
     saying
    -0.86
    atorship
    -0.85
     junto
    -0.84
    POSITIVE LOGITS
     japonais
    1.11
    ihr
    0.96
    λών
    0.89
    mawia
    0.89
     japonaise
    0.88
    damento
    0.88
     chinoise
    0.88
     INSERT
    0.88
     jepang
    0.88
     sırasında
    0.87
    Activations Density 0.033%

    No Known Activations