© 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. 118108
    Prev
    Next
    INDEX
    Explanations

    mystical ladder writings

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

    The neuron strongly activates on Spanish-language tokens (especially common function words like “la,” “de,” “y,” etc.), effectively detecting passages written in Spanish.

    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
     news
    -1.10
     Hector
    -0.83
    news
    -0.82
     ident
    -0.82
     Ife
    -0.81
     Priester
    -0.81
     Doge
    -0.80
     Satoshi
    -0.77
     ambitions
    -0.77
    NEWS
    -0.77
    POSITIVE LOGITS
     mys
    0.97
     Mys
    0.91
    Ladder
    0.91
     ladder
    0.90
     anonyme
    0.88
     cloud
    0.84
     Ladder
    0.84
     contemplative
    0.82
     desert
    0.81
    Mys
    0.80
    Activations Density 0.013%

    No Known Activations