© 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. 67296
    Prev
    Next
    INDEX
    Explanations

    question and answer counts

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

    The neuron strongly activates on numeric tokens (counts, IDs, statistics, etc.).

    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
    滋味
    -1.15
     from
    -1.11
     Homme
    -1.00
     antonio
    -0.99
     Kanada
    -0.95
     giovanni
    -0.95
     witcher
    -0.94
     ronaldo
    -0.94
     gmt
    -0.94
     for
    -0.93
    POSITIVE LOGITS
     answers
    1.45
     questions
    1.43
     discussion
    1.41
     pregunt
    1.40
     question
    1.31
     asking
    1.28
     clô
    1.22
     answer
    1.20
    这个问题
    1.18
    γρά
    1.18
    Activations Density 0.007%

    No Known Activations