© 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. 22-GEMMASCOPE-RES-131K
    7. 73421
    Prev
    Next
    INDEX
    Explanations

    its subsidiaries and affiliates

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

    The neuron detects legal or corporate “and its…” or “and/or its…” phrases that name a company’s subsidiaries, affiliates, divisions, or related entities.

    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
     about
    -1.39
     that
    -1.10
     needed
    -1.09
     for
    -1.05
     even
    -1.03
     estas
    -1.02
     Zugang
    -0.96
     esas
    -0.95
     while
    -0.93
    delige
    -0.93
    POSITIVE LOGITS
     who
    1.34
     employees
    1.29
     its
    1.19
     tts
    1.17
     affiliates
    1.02
    bės
    1.02
     ita
    1.01
    haraan
    1.00
    setzungen
    0.99
    フィシャル
    0.99
    Activations Density 0.030%

    No Known Activations