© 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. 69209
    Prev
    Next
    INDEX
    Explanations

    closing block comments

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

    This neuron activates specifically on numeric literal tokens (numbers) in the source 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_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
    -1.88
    雠
    -1.85
     cadeira
    -1.85
     fuis
    -1.80
     this
    -1.79
     "")
    -1.79
     these
    -1.77
    ۶
    -1.70
    瑧
    -1.69
     parede
    -1.66
    POSITIVE LOGITS
     
    2.19
     "
    1.90
     terutama
    1.89
    </h3>
    1.88
     -
    1.74
     serán
    1.66
    2
    1.63
     originally
    1.62
    niere
    1.59
     misalnya
    1.58
    Activations Density 0.217%

    No Known Activations