© 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. 64044
    Prev
    Next
    INDEX
    Explanations

    law firm, architectural firm, accounting firm

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

    The neuron is detecting mentions of professional firm names—especially law and accounting firms (e.g. “Toronto law firm … LLP,” “Frost, Briggs & Chamberlain,” “accounting firm Deloitte,” etc.).

    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
     компания
    -0.93
    olé
    -0.81
    пля
    -0.75
    ={({
    -0.74
    yak
    -0.73
     společnost
    -0.72
    現場
    -0.72
    董事
    -0.71
    Corp
    -0.71
     куби
    -0.70
    POSITIVE LOGITS
     firm
    4.84
     firms
    4.13
    firm
    3.47
     Firm
    3.45
    Firm
    3.28
     Firms
    2.91
     FIRM
    2.44
     firma
    2.34
     firmness
    2.33
     firme
    2.25
    Activations Density 0.080%

    No Known Activations