© 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. 79808
    Prev
    Next
    INDEX
    Explanations

    parameters, dimension, size, number

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

    The neuron fires on function words (like when, in, on, both, same) that introduce quantitative conditions or dependencies in mathematical or scientific statements.

    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
     denaro
    -1.09
    titure
    -1.06
     ↓,
    -1.05
    umumkan
    -1.05
    aigu
    -1.02
     teritor
    -1.02
     Speisen
    -1.02
    ый
    -1.01
     Probably
    -1.01
    🆗
    -0.99
    POSITIVE LOGITS
     size
    1.10
     design
    1.02
     $
    1.02
     designs
    1.00
    nischer
    0.99
    รวจ
    0.96
    จำนวน
    0.95
    bióticos
    0.94
     \
    0.94
     number
    0.93
    Activations Density 0.036%

    No Known Activations