© 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. 12721
    Prev
    Next
    INDEX
    Explanations

    great day or brilliant time

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

    The neuron sharply activates on the first few content words (and punctuation-adjacent tokens) at the start of each text segment.

    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
    咣
    -1.40
    呟
    -1.34
     inconsciente
    -1.31
     lü
    -1.30
    LayoutPanel
    -1.27
     PARKING
    -1.23
    情報は
    -1.23
     dokter
    -1.23
    ication
    -1.22
    嘀
    -1.22
    POSITIVE LOGITS
     through
    1.60
     from
    1.59
     now
    1.56
     for
    1.41
     them
    1.39
     out
    1.36
     about
    1.32
     or
    1.32
     like
    1.31
     started
    1.27
    Activations Density 0.209%

    No Known Activations