© 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. 2153
    Prev
    Next
    INDEX
    Explanations

    generated code or technical terms

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

    This neuron fires on low-frequency or “rare” tokens—e.g. numbers, names, technical terms, headings, and other uncommon words.

    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
    prg
    -0.71
     soggetti
    -0.68
     activates
    -0.68
     septembrie
    -0.67
    严格
    -0.66
     monitoring
    -0.66
     welcoming
    -0.66
     sicuramente
    -0.65
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    -0.65
    HBO
    -0.65
    POSITIVE LOGITS
     bathrooms
    0.77
     itinerary
    0.73
    おめでとう
    0.71
    ebles
    0.71
     Bathrooms
    0.70
    Bathrooms
    0.66
     dump
    0.66
     Eqs
    0.65
     ванной
    0.65
     toilets
    0.65
    Activations Density 0.089%

    No Known Activations