© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26787
    Prev
    Next
    INDEX
    Explanations

    splitting lists into parts

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

    The neuron fires most strongly on numeric values, units, and other technical terms that specify sizes, quantities, or measurements.

    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
     provides
    -0.84
     both
    -0.83
     rubbed
    -0.80
    đen
    -0.77
     emerges
    -0.76
    ällor
    -0.74
     providing
    -0.74
    ‌پدی
    -0.73
     provide
    -0.73
     establishing
    -0.73
    POSITIVE LOGITS
     split
    1.43
     divided
    1.31
    split
    1.30
     divid
    1.25
     segmented
    1.22
     staggered
    1.16
    Split
    1.16
    chun
    1.15
    分割
    1.14
     batches
    1.13
    Activations Density 0.093%

    No Known Activations