© 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. 105242
    Prev
    Next
    INDEX
    Explanations

    Well followed by introductory words

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

    The neuron strongly activates on the discourse‐marker “Well,” especially when it begins a sentence or clause.

    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
    -2.11
    ↵
    -1.87
     is
    -1.83
    }
    -1.81
     such
    -1.66
     måde
    -1.63
     調査
    -1.55
    锱
    -1.52
     たん
    -1.51
     すぐ
    -1.49
    POSITIVE LOGITS
    ╽
    1.98
    刧
    1.98
    ༷
    1.73
     mohou
    1.70
    withCredentials
    1.70
     increíble
    1.70
     aktivi
    1.69
    ଝ
    1.68
     позволит
    1.63
    檠
    1.63
    Activations Density 0.009%

    No Known Activations