© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 61321
    Prev
    Next
    INDEX
    Explanations

    using specific tools or code

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

    The neuron fires on the Russian instrumental-case phrase “помощью” (i.e. “with the help of/using”), marking method-of-action expressions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     has
    -1.97
    在于
    -1.91
     animados
    -1.89
     be
    -1.87
     $
    -1.84
     if
    -1.84
     contributes
    -1.82
    してます
    -1.77
     magnética
    -1.73
     was
    -1.73
    POSITIVE LOGITS
    ляции
    2.20
    颪
    2.19
     our
    2.08
     juſ
    2.05
     mahu
    2.03
     イラストレーション
    2.02
     competen
    1.99
     dezelve
    1.95
     Tengo
    1.95
    臵
    1.89
    Activations Density 0.006%

    No Known Activations