© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 78594
    Prev
    Next
    INDEX
    Explanations

    "thus", "therefore", "so"

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    сный
    -0.86
    ộc
    -0.80
     Partly
    -0.79
    ESPEC
    -0.77
     které
    -0.77
     भी
    -0.77
     ด้วย
    -0.76
    partly
    -0.76
    preferably
    -0.75
     Agg
    -0.75
    POSITIVE LOGITS
     therefore
    3.97
     donc
    3.31
     quindi
    3.19
     więc
    3.05
     thus
    2.84
     dunque
    2.70
    Therefore
    2.64
     Therefore
    2.52
    therefore
    2.48
    所以
    2.44
    Activations Density 0.010%

    No Known Activations