© 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. 10-GEMMASCOPE-RES-131K
    7. 120121
    Prev
    Next
    INDEX
    Explanations

    temporal markers or specific instances

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

    The neuron fires on nouns and noun phrases that appear in quantitative or measurement contexts (e.g. “time period,” “day,” “occasion,” “fees,” “problem,” “case,” “relationship,” “platforms”), especially around 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_10/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
     our
    -1.77
     you
    -1.24
     noastre
    -1.18
     bym
    -1.17
    我和
    -1.17
     mountainous
    -1.11
    まく
    -1.11
     because
    -1.11
     différence
    -1.10
     και
    -1.09
    POSITIVE LOGITS
    ňte
    1.16
     منذ
    1.16
     вместо
    1.13
     than
    1.13
     свои
    1.08
    cadilly
    1.07
    бира
    1.02
     gratuite
    1.01
     incluse
    1.00
    }$_
    1.00
    Activations Density 0.395%

    No Known Activations