© 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. 22-GEMMASCOPE-RES-131K
    7. 95838
    Prev
    Next
    INDEX
    Explanations

    edge ruler

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

    The neuron flags uncommon, content-bearing tokens—especially domain-specific technical terms or other key nouns—throughout 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_22/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
    tldr
    -1.08
     direta
    -0.99
     rurales
    -0.96
    ఎ
    -0.93
     tropicales
    -0.91
    ところに
    -0.91
     NPCs
    -0.90
     orm
    -0.90
    SUNDAY
    -0.89
    ។
    -0.89
    POSITIVE LOGITS
     of
    1.27
     diseñada
    0.91
     idea
    0.90
    ویزی
    0.88
     framework
    0.86
    いただきます
    0.82
     feature
    0.80
    valy
    0.79
    scaleY
    0.79
    HAUS
    0.79
    Activations Density 0.160%

    No Known Activations