© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7246
    Prev
    Next
    INDEX
    Explanations

    os / path

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

    This neuron is highly activated by the two-letter sequence “Os” (an uppercase O followed by s), as found at the start of proper-name tokens like Osborne, Osram, Osceola, Oshawa, etc.

    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
    仪表
    0.55
     sécrétion
    0.55
    encountered
    0.55
    জ্জনক
    0.54
    带来了
    0.53
    quetas
    0.53
     化粧
    0.52
    惮
    0.52
     Inicio
    0.52
    ሱን
    0.52
    POSITIVE LOGITS
     Os
    0.93
    Os
    0.80
     Osm
    0.79
     os
    0.75
     Ос
    0.69
     Osbourne
    0.69
    Osm
    0.67
     Osborne
    0.64
     osm
    0.64
     Oss
    0.61
    Activations Density 0.113%

    No Known Activations