© 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. 67909
    Prev
    Next
    INDEX
    Explanations

    detecting and predicting methods

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

    The main thing this neuron does is spot technical descriptions of statistical or computational modeling methods (e.g. terms like “model,” “analysis,” “distribution,” “function,” “combines,” etc.) in scientific 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_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
     excepcional
    -1.21
    Aktu
    -1.16
    אם
    -1.10
    skrä
    -1.09
     membuk
    -1.06
     bahagia
    -1.05
     chuyến
    -1.05
     häufig
    -1.05
    Generally
    -1.05
     السياس
    -1.04
    POSITIVE LOGITS
    1.24
    也能
    1.23
     permite
    1.17
     elegantly
    1.16
     eikä
    1.12
     และ
    1.11
    StringBuffer
    1.10
    在意
    1.10
     ו
    1.09
    peut
    1.09
    Activations Density 0.192%
    No Known Activations