© 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. 34-GEMMASCOPE-RES-131K
    7. 92596
    Prev
    Next
    INDEX
    Explanations

    instead, while, without

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

    The main thing this neuron does is detect references to quantitative measurement or metrics and related analytic 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_34/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
     honest
    -1.18
     was
    -1.16
    也能
    -1.13
     after
    -1.12
     trustworthy
    -1.06
     Schwierigkeiten
    -1.06
     gradual
    -1.05
    も行
    -1.05
     liksom
    -1.05
     schwerer
    -1.05
    POSITIVE LOGITS
     while
    2.05
     instead
    1.75
     without
    1.64
     вместо
    1.59
     terwijl
    1.58
     whilst
    1.55
     mientras
    1.51
     mentre
    1.44
    instead
    1.41
    while
    1.38
    Activations Density 0.095%

    No Known Activations