© 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. 13097
    Prev
    Next
    INDEX
    Explanations

    reporting sources' names

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

    The main thing this neuron does is detect numeric tokens (numbers, counts, percentages, and other digit sequences).

    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
    얻
    -0.85
    OFFER
    -0.83
    잘
    -0.82
    𝗬
    -0.82
    而后
    -0.82
    会社の
    -0.80
     griega
    -0.78
     cadrul
    -0.78
     vente
    -0.77
     veliko
    -0.77
    POSITIVE LOGITS
     jednod
    1.03
     interni
    0.97
    НИ
    0.97
     trö
    0.94
    MenuInflater
    0.94
    сюда
    0.93
     stava
    0.92
     voleva
    0.92
     своему
    0.92
    PERTY
    0.91
    Activations Density 0.011%

    No Known Activations