© 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. 106584
    Prev
    Next
    INDEX
    Explanations

    tax loopholes and evasion

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

    The neuron strongly activates on terminology related to tax planning—especially references to “tax shelter,” “taxation,” and similar tax‐avoidance or shelter concepts.

    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
     prezzo
    -0.79
     months
    -0.78
    jamento
    -0.77
    withCredentials
    -0.72
    nungszeiten
    -0.72
     anello
    -0.71
    हर
    -0.70
    汞
    -0.70
    estroy
    -0.68
     suffered
    -0.68
    POSITIVE LOGITS
     avoidance
    1.52
     loopholes
    1.34
     evasion
    1.17
     dodging
    1.16
     minimization
    1.13
     evade
    1.13
     eva
    1.11
     Avoidance
    1.10
    aggressive
    1.09
     strategies
    1.05
    Activations Density 0.026%

    No Known Activations