© 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. 10-GEMMASCOPE-RES-131K
    4. 42572
    Prev
    Next
    INDEX
    Explanations

    Spanish and Portuguese prefixes + suffixes

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

    The neuron fires on word‐internal stem prefixes—i.e. the beginnings of longer content words (common morpheme stems) across languages.

    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
    だけです
    -1.88
    までは
    -1.84
    着一
    -1.84
    を持つ
    -1.84
    ↵
    -1.84
    ppuden
    -1.83
    while
    -1.82
     llegamos
    -1.81
    家里
    -1.80
     potente
    -1.77
    POSITIVE LOGITS
     and
    2.64
    8
    2.13
     podoba
    1.96
    1
    1.89
     большим
    1.87
     zosta
    1.85
    끊
    1.82
     jika
    1.81
     errori
    1.79
     untuk
    1.76
    Activations Density 0.023%

    No Known Activations