© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 92900
    Prev
    Next
    INDEX
    Explanations

    was + followed by action

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

    This neuron detects past-progressive verb constructions (e.g. “was putting,” “was walking,” “were going”)

    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
     потому
    -0.93
     betrayal
    -0.90
     tarvit
    -0.83
     avoient
    -0.82
     algebras
    -0.82
     ведь
    -0.82
    tivation
    -0.82
     isolamento
    -0.82
     女人
    -0.81
     raiſ
    -0.81
    POSITIVE LOGITS
    ícias
    1.04
    </tfoot>
    1.02
    ward
    0.97
    러한
    0.92
    Almost
    0.91
     think
    0.89
     focusing
    0.86
     waiting
    0.86
     begy
    0.86
     awaiting
    0.85
    Activations Density 0.014%

    No Known Activations