© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 47992
    Prev
    Next
    INDEX
    Explanations

    platform or platforms

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

    The neuron strongly activates on the structural metadata and section‐heading terminology (e.g. station names, “Lines,” “Station layout,” distances, platform counts) typical of railway station infoboxes and descriptions.

    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
    vikle
    -0.71
    ariales
    -0.68
     financières
    -0.68
    regs
    -0.66
     induct
    -0.65
    джет
    -0.64
    logico
    -0.64
     biryani
    -0.64
     centric
    -0.64
     לי
    -0.64
    POSITIVE LOGITS
     platform
    2.58
    platform
    2.33
     platforms
    2.25
    Platform
    2.23
     Platform
    2.19
     plataforma
    1.95
     Platforms
    1.88
     платфор
    1.84
    platforms
    1.78
    PLATFORM
    1.71
    Activations Density 0.036%

    No Known Activations