© 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
    2. Gemma-3-27B-IT
    3. 13-GEMMASCOPE-2-TRANSCODER-262K
    4. 42785
    Prev
    Next
    INDEX
    Explanations

    I'm currently processing the provided lists to identify a consistent pattern or theme that describes the neuron's behavior. My goal is to find a succinct phrase (3-20 words) that encapsulates this behavior, focusing on what the neuron might be detecting or predicting in the input data. I'll be looking for recurring concepts, syntactic structures, or word sequences across the different lists, especially `TOP_POSITIVE_LOGITS` and `TOP_ACTIVATING_TEXTS`, while considering `MAX_ACTIVATING_TOKENS` and `TOKENS_AFTER_MAX_ACTIVATING_TOKEN` for contextual clues about how these patterns manifest.**Explanation of Neuron Behavior:**multi-language context indicators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_13_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     полу
    0.47
     cesium
    0.47
    ServiceGroupID
    0.46
    ttä
    0.45
     Internationale
    0.45
     gå
    0.45
     研究
    0.44
     മുത
    0.44
     bhuv
    0.44
     Wohl
    0.43
    POSITIVE LOGITS
    மை
    0.50
    <
    0.49
    लेन
    0.48
     ▲
    0.47
    गण
    0.47
    विल
    0.47
    льнай
    0.47
    ларда
    0.46
    ларни
    0.46
     Acuerdo
    0.46
    Activations Density 0.001%

    No Known Activations