© 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. 66491
    Prev
    Next
    INDEX
    Explanations

    data analysis

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

    This neuron activates on tokens referring to scientific data or measurements—e.g. dataset sizes, imaging modalities (CT, US), scans, and related numerical‐data terms.

    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
     and
    -1.21
    ()/
    -1.04
     fabricar
    -1.01
    дельник
    -0.99
    6
    -0.97
     as
    -0.95
     all
    -0.94
     when
    -0.93
    (),
    -0.93
     vinculado
    -0.93
    POSITIVE LOGITS
     récemment
    1.45
     from
    1.23
     of
    1.21
    ⸻
    1.13
     sœurs
    1.12
     udang
    1.10
     ActionListener
    1.09
     Jadi
    1.09
     illimit
    1.08
     Як
    1.07
    Activations Density 0.188%

    No Known Activations