© 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. 58642
    Prev
    Next
    INDEX
    Explanations

    actions or transitions

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

    The neuron responds strongly to past-tense verbs (words ending in or representing actions completed in the past).

    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
     in
    -2.38
     for
    -1.70
     only
    -1.69
     also
    -1.66
    )$.
    -1.64
    ,
    -1.64
     Verkä
    -1.53
    What
    -1.48
     gewünschten
    -1.47
    ぶりの
    -1.44
    POSITIVE LOGITS
     maksi
    1.66
    стон
    1.64
    杙
    1.63
    謌
    1.50
    ilirsiniz
    1.48
     our
    1.45
    STILL
    1.41
     venne
    1.41
     Meksiko
    1.39
     Теперь
    1.38
    Activations Density 0.086%

    No Known Activations