© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35436
    Prev
    Next
    INDEX
    Explanations

    have been ... since

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

    The neuron strongly responds to verbs in the perfect‐progressive or continuous aspect (e.g. “have/been …-ing”)—i.e. ongoing action phrases.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    -1.88
     first
    -1.63
     ancora
    -1.22
     one
    -1.21
     первая
    -1.09
     comune
    -1.09
     erster
    -1.09
     einer
    -1.09
     being
    -1.06
     deux
    -1.06
    POSITIVE LOGITS
     since
    3.13
     lately
    2.33
    since
    2.23
     SINCE
    1.94
     منذ
    1.94
     depuis
    1.84
     sejak
    1.73
     Lately
    1.59
    Since
    1.51
    Seit
    1.49
    Activations Density 0.050%

    No Known Activations