© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38797
    Prev
    Next
    INDEX
    Explanations

    but followed by continuation

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

    This neuron strongly activates on the contrastive conjunction “but.”

    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
    に
    -1.13
    」。
    -1.01
    が
    -1.01
    で
    -1.01
    Література
    -0.98
     "")
    -0.95
     به‌عنوان
    -0.93
    肖像
    -0.92
    "}"
    -0.92
    ).(
    -0.91
    POSITIVE LOGITS
    AfterViewInit
    1.00
    事で
    0.94
     alimentare
    0.92
    になれば
    0.92
     ibang
    0.92
    也就
    0.89
    partic
    0.89
     破解
    0.88
     caratteri
    0.88
     procra
    0.85
    Activations Density 0.039%

    No Known Activations