© 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. 104444
    Prev
    Next
    INDEX
    Explanations

    parts of speech and descriptions

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

    The neuron is essentially a “content‐word” detector: it fires on substantive tokens (nouns, proper names, numbers, technical terms) and stays off for common function 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_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
     abgerufen
    -1.03
     ``
    -1.02
    tad
    -0.95
    corresponding
    -0.91
     corresponding
    -0.91
     "_
    -0.91
     производителя
    -0.91
     lông
    -0.89
    Email
    -0.89
     †
    -0.88
    POSITIVE LOGITS
    尤其
    1.01
    Absol
    0.89
    ticularly
    0.88
    特に
    0.87
     granito
    0.85
    cfr
    0.84
    CARBON
    0.82
    Bjs
    0.82
     marat
    0.82
     comien
    0.81
    Activations Density 0.065%

    No Known Activations