© 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. 31405
    Prev
    Next
    INDEX
    Explanations

    words followed by specific prepositions or punctuation

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

    This neuron responds to promotional or marketing language—high-value adjectives and sales-pitch phrases describing products, services, quality, and expertise.

    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
     without
    -1.02
     if
    -0.87
    してしまった
    -0.85
     while
    -0.84
     cleverly
    -0.82
     gets
    -0.82
     undeniably
    -0.82
     supremely
    -0.82
     took
    -0.81
    媖
    -0.81
    POSITIVE LOGITS
     distinctive
    1.10
    ousy
    1.09
     fantastic
    1.08
     gorgeous
    1.07
     magnificent
    1.05
     tinha
    1.03
     superstar
    1.01
     breathtaking
    1.00
     vibrant
    0.99
     provavelmente
    0.98
    Activations Density 0.012%

    No Known Activations