© 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. 102631
    Prev
    Next
    INDEX
    Explanations

    gap and disparity

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

    The main thing this neuron does is detect words that signal a quantitative difference or disparity (e.g. “disparity,” “gap,” “difference,” “discrepancy”).

    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
    ванта
    -0.82
     induk
    -0.80
    缚
    -0.79
    LowerCase
    -0.79
     byli
    -0.78
     ImageIcon
    -0.75
     citar
    -0.75
     complementa
    -0.74
    padu
    -0.73
    Trung
    -0.73
    POSITIVE LOGITS
     widening
    2.38
     widened
    2.23
     gap
    2.13
     widen
    2.05
     widest
    1.95
     chasm
    1.85
     yawning
    1.73
    wid
    1.73
    gap
    1.63
     Gap
    1.60
    Activations Density 0.039%

    No Known Activations