© 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. 10-GEMMASCOPE-RES-131K
    7. 101943
    Prev
    Next
    INDEX
    Explanations

    nevertheless, nonetheless, however

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

    The neuron strongly activates on contrastive discourse markers (e.g. “Nevertheless,” “However,” etc.).

    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
     and
    -2.06
    i
    -1.93
    E
    -1.90
     At
    -1.76
    At
    -1.74
    ;
    -1.70
     from
    -1.70
     evidently
    -1.63
     enormous
    -1.60
     using
    -1.58
    POSITIVE LOGITS
    媄
    2.27
    ſelf
    2.13
     verdaderos
    2.09
     ホルダー
    2.08
     ganadores
    2.06
     tranſ
    2.05
     nacidos
    2.05
     verhind
    2.03
    橚
    2.02
    ︲
    2.02
    Activations Density 0.022%

    No Known Activations