© 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. 34-GEMMASCOPE-RES-131K
    7. 60153
    Prev
    Next
    INDEX
    Explanations

    numbers and dates

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

    The neuron is essentially detecting numeric tokens (digits and 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
     analyzed
    -1.70
     остается
    -1.62
     meager
    -1.54
     favorite
    -1.53
     favorably
    -1.52
     еще
    -1.51
     neighbors
    -1.51
     neighborhoods
    -1.50
     organized
    -1.49
     придется
    -1.49
    POSITIVE LOGITS
    🏟
    1.65
     бампер
    1.59
    Whilst
    1.46
     whilst
    1.43
     Sabtu
    1.37
     tavsiye
    1.32
     UK
    1.30
     塗装
    1.29
     Tahun
    1.27
     cilt
    1.26
    Activations Density 0.088%

    No Known Activations