© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 130833
    Prev
    Next
    INDEX
    Explanations

    This neuron detects verbs in the past tense (both regular ‑ed forms and common irregular past-tense verbs).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    verbs followed by direction

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     which
    -1.71
     or
    -1.68
     any
    -1.46
     there
    -1.38
     what
    -1.34
     whose
    -1.34
     ....
    -1.34
     if
    -1.27
     just
    -1.24
     people
    -1.23
    POSITIVE LOGITS
     then
    2.53
     vervolgens
    1.55
     inoltre
    1.52
    Then
    1.41
     alcuni
    1.38
     まる
    1.34
     entonces
    1.32
     Then
    1.32
     então
    1.29
    then
    1.28
    Activations Density 0.058%

    No Known Activations