© 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. 3756
    Prev
    Next
    INDEX
    Explanations

    but introducing a contrast

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

    The neuron strongly activates on the conjunction “but,” i.e. it’s detecting the contrastive “but” in sentences.

    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
     in
    -1.23
    гата
    -1.15
    Siti
    -1.10
     vilja
    -1.09
    </u>
    -1.09
    蜮
    -1.06
    </i>
    -1.06
    vell
    -1.03
    用戶
    -1.03
     guste
    -1.02
    POSITIVE LOGITS
    鷙
    1.32
    埽
    1.30
     produttore
    1.24
    karni
    1.23
    vaient
    1.21
    mathbf
    1.17
    ⤙
    1.16
     spetta
    1.12
     whom
    1.10
    cantara
    1.10
    Activations Density 0.019%

    No Known Activations