© 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. 34-GEMMASCOPE-RES-131K
    4. 92255
    Prev
    Next
    INDEX
    Explanations

    prevention and protection

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

    technical or scientific terms in specialized documents or professional literature.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron responds to long, domain-specific words—especially technical or scientific terms with many syllables.

    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
    letoe
    -0.88
     unprotected
    -0.81
    svr
    -0.79
    delser
    -0.78
    ueuse
    -0.76
    懐かしい
    -0.76
    agram
    -0.75
    预期
    -0.74
    ocentric
    -0.74
    itz
    -0.73
    POSITIVE LOGITS
     anti
    1.91
     disease
    1.48
     pollution
    1.44
     анти
    1.38
     acne
    1.35
     crime
    1.34
     earthquake
    1.33
     injury
    1.30
     disaster
    1.28
     accident
    1.24
    Activations Density 0.253%

    No Known Activations