© 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. 13782
    Prev
    Next
    INDEX
    Explanations

    data transformation and processing

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

    The neuron fires on technical task or process terms—single-word or compound nouns that name programming or data-processing steps (e.g. “processing,” “pipeline,” “wrangling,” “normalization”).

    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
    butuhkan
    -1.12
     esquí
    -1.09
    乁
    -1.02
    いましたが
    -1.01
    sebagai
    -1.00
    devtools
    -0.99
     ciclismo
    -0.96
    
    -0.94
    แทน
    -0.94
    Farma
    -0.93
    POSITIVE LOGITS
     transformations
    1.41
    处理
    1.38
     transformation
    1.28
     processing
    1.24
     to
    1.17
     transform
    1.16
     treatment
    1.15
    Transform
    1.13
    transform
    1.13
    加工
    1.13
    Activations Density 0.133%

    No Known Activations