© 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. 36403
    Prev
    Next
    INDEX
    Explanations

    actions or future events

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

    The neuron fires on the primary action verbs (especially continuous and simple present forms) that drive a sentence’s main events.

    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
     and
    -1.20
     began
    -1.09
     successfully
    -1.09
     recently
    -1.03
     located
    -1.02
     had
    -1.02
    還在
    -0.98
     rũ
    -0.97
     still
    -0.96
     jardines
    -0.95
    POSITIVE LOGITS
     will
    2.05
     tomorrow
    1.38
     gonna
    1.11
     to
    1.09
     bakal
    1.08
    ที่จะ
    1.01
    tungen
    1.01
     rádio
    1.00
     будет
    1.00
     kuris
    0.99
    Activations Density 0.058%

    No Known Activations