© 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. 34-GEMMASCOPE-RES-131K
    4. 82256
    Prev
    Next
    INDEX
    Explanations

    units of measurement

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

    The main thing this neuron does is detect numerical quantity tokens (e.g. multi-digit numbers) in the 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_34/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
     крупных
    -0.75
    terminal
    -0.74
     $.
    -0.72
     exploited
    -0.71
     wprowad
    -0.70
     jene
    -0.69
     warehouse
    -0.69
    katie
    -0.68
     хорошего
    -0.68
    >.
    -0.68
    POSITIVE LOGITS
     tonnes
    1.09
     barrels
    1.07
     ton
    0.96
     tonne
    0.95
     tons
    0.88
    Zna
    0.87
    jeron
    0.85
    Ton
    0.84
     BOL
    0.84
     FEM
    0.84
    Activations Density 0.019%

    No Known Activations