© 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. 75427
    Prev
    Next
    INDEX
    Explanations

    us or we

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

    This neuron fires on capitalized or numeral tokens—i.e. words that start sentences (or are proper nouns) and numbers.

    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
     Ajay
    -0.74
    getDoctrine
    -0.71
    αι
    -0.71
    pshire
    -0.68
    irmingham
    -0.67
     digo
    -0.67
    altura
    -0.66
     Ordens
    -0.66
    ֶה
    -0.66
    Opts
    -0.65
    POSITIVE LOGITS
     we
    3.84
     We
    3.73
    We
    3.22
    we
    2.83
    我们
    2.47
     WE
    2.44
    我們
    2.38
     мы
    2.34
     我们
    2.17
    WE
    2.08
    Activations Density 0.147%

    No Known Activations