© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66912
    Prev
    Next
    INDEX
    Explanations

    statement beginnings

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

    The neuron detects the core subject–verb pair of a main clause.

    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
     после
    -1.13
     poprzez
    -1.12
    雖
    -1.02
    ｻ
    -0.99
     pregunt
    -0.98
     empieza
    -0.98
     aún
    -0.97
     aufmerksam
    -0.96
     setelah
    -0.94
     mientras
    -0.93
    POSITIVE LOGITS
    _
    1.31
     poème
    1.18
     kalori
    1.16
     bambú
    1.13
    ῖν
    1.12
     montant
    1.11
     bénévol
    1.09
     mostaza
    1.05
     сестра
    1.05
    🫃
    1.05
    Activations Density 0.007%

    No Known Activations