© 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. 10-GEMMASCOPE-RES-131K
    4. 19078
    Prev
    Next
    INDEX
    Explanations

    systems, series, data, speech, chemical

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

    technical and scientific terms across different fields including data systems, time series, power systems, and quantum physics.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on content‐heavy scientific and technical terms—especially multiword domain keywords (e.g. “power systems,” “time series,” “quantum systems,” “multivariate data,” etc.)—in paper titles and abstracts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     they
    -1.52
    还得
    -1.48
    出来ました
    -1.48
     possíveis
    -1.46
    我这
    -1.45
     твои
    -1.45
    几分
    -1.38
    赑
    -1.38
    牍
    -1.38
     accepted
    -1.38
    POSITIVE LOGITS
    standigheden
    1.46
    :
    1.34
    zku
    1.29
    Leírás
    1.23
     információk
    1.21
     роки
    1.20
    strtolower
    1.20
    яз
    1.20
    Sow
    1.19
    Reti
    1.18
    Activations Density 0.852%

    No Known Activations