© 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. 8758
    Prev
    Next
    INDEX
    Explanations

    JSON, HTTP, AI, GU

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

    This neuron activates on all-caps abbreviations and short uppercase codes—e.g. time markers (AM, UTC), file/media formats (MP3, CD, DVD), units or specs (HP), and acronyms or state codes (TV, KY).

    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
    E
    0.86
    A
    0.75
    SIC
    0.69
    EPA
    0.68
    EH
    0.66
    SIG
    0.65
    ECC
    0.65
     एफ
    0.64
    Ao
    0.64
    S
    0.64
    POSITIVE LOGITS
    aaS
    0.78
     ಯ
    0.70
    يته
    0.66
    toJSON
    0.65
     UInt
    0.65
     minta
    0.65
    egal
    0.64
    कैसे
    0.62
     combustibles
    0.62
    COc
    0.62
    Activations Density 0.688%

    No Known Activations