© 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. 67002
    Prev
    Next
    INDEX
    Explanations

    descriptive adjective noun pairs

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

    The main thing this neuron does is spot promotional or marketing hype language—especially superlative quality claims and persuasive product adjectives.

    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
    就可以了
    -1.09
     However
    -1.05
     tuttavia
    -1.04
     però
    -1.00
    ADVANTAGES
    -0.98
     however
    -0.98
    似乎
    -0.98
    是我的
    -0.97
    〜！
    -0.96
    Zusammenfassung
    -0.96
    POSITIVE LOGITS
     leurs
    1.05
     Sirius
    1.03
    ListTile
    0.92
    ᴛ
    0.92
     الدوري
    0.91
    PhysRev
    0.91
    ミ
    0.90
     again
    0.89
    espn
    0.89
     stanovnika
    0.88
    Activations Density 0.022%

    No Known Activations