© 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. 14543
    Prev
    Next
    INDEX
    Explanations

    recommend to friends and family

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

    The neuron fires on words and phrases that express recommending or referring a product/service (e.g. “recommend,” “definitely refer,” “will recommend to all friends and family”).

    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
     хуже
    -0.99
     letzten
    -0.88
    Compare
    -0.88
     or
    -0.83
     anticipation
    -0.82
    сторо
    -0.81
    Estimate
    -0.81
     onus
    -0.81
     keep
    -0.80
    ↵↵↵↵↵
    -0.79
    POSITIVE LOGITS
     friends
    1.34
     anyone
    1.28
     colleagues
    1.21
     others
    1.10
     anybody
    1.06
     potensi
    1.05
     Clientes
    1.04
    ೢ
    1.03
    笕
    1.02
    wohnt
    1.02
    Activations Density 0.003%

    No Known Activations