© 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. 22-GEMMASCOPE-RES-131K
    7. 13286
    Prev
    Next
    INDEX
    Explanations

    years followed by he

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

    The main thing this neuron does is detect numeric year mentions (four-digit dates) in the text.

    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
     that
    -1.34
    introdu
    -1.10
    λοι
    -1.09
     appris
    -1.06
     about
    -1.06
    many
    -1.05
     some
    -1.05
    で見る
    -1.03
    redu
    -1.02
    about
    -1.01
    POSITIVE LOGITS
     hän
    1.34
     she
    1.21
     you
    1.15
     he
    1.13
     when
    1.07
     عندما
    1.06
    他
    1.02
    undan
    1.00
     वह
    0.98
    тири
    0.97
    Activations Density 0.041%

    No Known Activations