© 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. 34-GEMMASCOPE-RES-131K
    7. 30391
    Prev
    Next
    INDEX
    Explanations

    previous events

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

    The neuron is picking out past‐tense regular verbs ending in “-ed.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    现在的
    -1.20
     nächsten
    -1.07
     Subsequent
    -0.96
     kolejne
    -0.95
     suivants
    -0.94
    现
    -0.93
     huidige
    -0.91
    現在の
    -0.91
     actuels
    -0.90
     بعدی
    -0.88
    POSITIVE LOGITS
     earlier
    8.56
    earlier
    6.28
    Earlier
    6.00
     Earlier
    5.56
     ранее
    4.28
     wcześniej
    4.19
     sebelumnya
    4.09
     previously
    4.00
     tidigare
    3.95
     previous
    3.84
    Activations Density 0.144%

    No Known Activations