© 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. 8341
    Prev
    Next
    INDEX
    Explanations

    been developing

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

    The neuron fires on perfect‐progressive verb phrases—that is, forms like “has/have been” followed by an “‐ing” verb.

    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
     still
    -1.46
     two
    -1.31
     wit
    -1.30
     pierwszy
    -1.23
     first
    -1.23
    langkah
    -1.23
     many
    -1.21
     there
    -1.21
     one
    -1.19
     that
    -1.17
    POSITIVE LOGITS
     lately
    1.54
    耵
    1.29
    orsese
    1.29
     marié
    1.27
     JTextField
    1.24
    fetera
    1.23
    最近
    1.23
     filosofia
    1.20
    ogaster
    1.20
     navigu
    1.19
    Activations Density 0.129%

    No Known Activations