© 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. 14429
    Prev
    Next
    INDEX
    Explanations

    invention relates to

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

    This neuron lights up strongly on common high-frequency function words (especially conjunctions and auxiliaries like “and,” “are,” and “did”) in patent-style text.

    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
     zove
    -1.30
     podjela
    -1.17
    depuis
    -1.14
     armée
    -1.11
     tén
    -1.09
     ralla
    -1.09
     elegantes
    -1.08
     ministra
    -1.08
    НЫ
    -1.06
     célè
    -1.05
    POSITIVE LOGITS
     will
    1.19
    b
    1.15
    ‘
    1.12
    text
    1.09
    p
    1.09
    h
    1.09
     this
    1.08
     that
    1.07
    ch
    1.07
    ab
    1.05
    Activations Density 0.050%

    No Known Activations