© 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-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 10343
    Prev
    Next
    INDEX
    Explanations

    platform-independent

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

    The neuron is highly specialized to detect the word “platform” (and its subword variants) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    лле
    0.71
    ми
    0.70
     llenar
    0.64
    חת
    0.63
    μη
    0.63
    reciation
    0.63
     रोहित
    0.61
    ৰ
    0.61
    üller
    0.60
     就是
    0.60
    POSITIVE LOGITS
     Platforms
    1.52
     platforms
    1.48
     Platform
    1.47
    Platform
    1.41
     platform
    1.41
    platform
    1.34
    平台的
    1.34
    平台
    1.32
    platforms
    1.32
     Plattform
    1.28
    Activations Density 0.064%

    No Known Activations