© 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. 130739
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens, i.e. explicit numbers or quantities in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    physical quantities and parameters

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    や
    -0.93
     christlichen
    -0.88
     assembly
    -0.86
     three
    -0.82
     Grüßen
    -0.81
    🚃
    -0.81
     only
    -0.81
    !!!!
    -0.80
    などの
    -0.79
    是最
    -0.79
    POSITIVE LOGITS
    感じが
    0.99
     różnych
    0.98
     sorteo
    0.96
    人たち
    0.92
    çant
    0.91
     relativement
    0.91
     wciąż
    0.91
    ところです
    0.90
    感じです
    0.89
    convertView
    0.89
    Activations Density 0.033%

    No Known Activations