© 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. 85040
    Prev
    Next
    INDEX
    Explanations

    activities and ongoing actions

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

    The neuron strongly activates on English verbs in their continuous/gerund (“‐ing”) form.

    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
     toujours
    -1.51
    tka
    -1.45
     artikel
    -1.36
    тна
    -1.36
     yar
    -1.32
    wiz
    -1.28
    LECTIONS
    -1.27
    ktion
    -1.26
     hanno
    -1.24
    okok
    -1.24
    POSITIVE LOGITS
     =
    1.58
    もちろん
    1.48
     außerhalb
    1.36
     určit
    1.30
     unpredictable
    1.30
     связан
    1.29
    カップル
    1.27
    Each
    1.26
    日曜日
    1.26
     $-
    1.19
    Activations Density 0.098%

    No Known Activations