© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 36020
    Prev
    Next
    INDEX
    Explanations

    defines or expresses formulas

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

    The neuron detects definitional or formula‐introduction phrases (e.g. “is defined as,” “as follows,” “the following formula,” “looks like,” “can be described as”).

    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
    所有人都
    -1.20
     Ã
    -1.11
    abinieri
    -1.10
     exceptionally
    -1.09
     alcune
    -1.06
     alcuni
    -1.06
    gier
    -1.05
     –
    -1.03
     ●
    -1.01
    になっている
    -1.01
    POSITIVE LOGITS
     with
    1.44
    🔽
    1.28
     unrelenting
    1.20
     after
    1.18
     without
    1.16
     manzanas
    1.16
     älskar
    1.15
    Following
    1.13
     playful
    1.13
     enticing
    1.13
    Activations Density 0.007%

    No Known Activations