© 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. 34-GEMMASCOPE-RES-131K
    7. 49748
    Prev
    Next
    INDEX
    Explanations

    need to / should / must

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

    This neuron fires on Cyrillic (Russian) text, detecting when the document contains Russian-language tokens.

    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
     می‌تواند
    -1.06
    viato
    -0.99
     strategis
    -0.96
    ✱
    -0.91
     pernic
    -0.90
    瓚
    -0.90
     понадобится
    -0.90
     justa
    -0.88
    continence
    -0.88
     أخرى
    -0.87
    POSITIVE LOGITS
     trzeba
    3.86
     нужно
    3.59
     należy
    3.20
     treba
    3.17
     should
    3.16
     надо
    3.00
     need
    2.95
     необходимо
    2.92
     следует
    2.81
     потрібно
    2.81
    Activations Density 0.022%

    No Known Activations