© 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. 105728
    Prev
    Next
    INDEX
    Explanations

    verbs indicating past actions

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

    The neuron fires on inflected verb forms (e.g. past‐tense “-ed” and third‐person present “-s/-es”) throughout the text.

    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
     soñar
    -1.56
     milyon
    -1.52
    ⦑
    -1.49
    guigu
    -1.48
    homonymie
    -1.45
     olvidar
    -1.44
    Składniki
    -1.42
     acogedor
    -1.39
     jepang
    -1.38
     motocicletas
    -1.38
    POSITIVE LOGITS
    in
    1.59
    is
    1.57
    en
    1.56
    ge
    1.55
    de
    1.55
    an
    1.53
    op
    1.47
    lat
    1.47
     his
    1.46
    another
    1.45
    Activations Density 0.157%

    No Known Activations