© 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. 2194
    Prev
    Next
    INDEX
    Explanations

    you can redistribute or modify

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

    The neuron fires on subword pieces of Latvian verb forms (e.g. in “paliks” and “liecina”), i.e. it detects inflected Latvian verbs.

    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
    ancers
    -0.93
    fordern
    -0.90
    äumt
    -0.87
    얽
    -0.86
    درمان
    -0.85
     Diputados
    -0.83
     Effekt
    -0.83
    ㉨
    -0.83
     Ministries
    -0.83
    begre
    -0.82
    POSITIVE LOGITS
     гру
    0.93
     is
    0.91
    ffler
    0.80
     Georgia
    0.80
     дви
    0.79
     enigma
    0.78
     избежать
    0.77
     subjecting
    0.75
     should
    0.75
     рады
    0.75
    Activations Density 0.000%

    No Known Activations