© 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. 10-GEMMASCOPE-RES-131K
    7. 71729
    Prev
    Next
    INDEX
    Explanations

    for noun phrases

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

    the presence of common English function words (especially prepositions and linking words such as “in,” “by,” “with,” “for,” “that,” 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
     refroid
    -1.73
    を追加
    -1.45
     rembour
    -1.41
    Nouvelle
    -1.41
     escribir
    -1.39
     déten
    -1.39
    lorem
    -1.38
     appré
    -1.35
     apporter
    -1.34
    铍
    -1.34
    POSITIVE LOGITS
     was
    1.96
     like
    1.56
     using
    1.52
     for
    1.52
     gave
    1.45
     most
    1.42
     â
    1.42
     more
    1.40
     on
    1.40
     a
    1.40
    Activations Density 0.179%

    No Known Activations