© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 101034
    Prev
    Next
    INDEX
    Explanations

    machine learning concepts

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

    technical terms and acronyms related to machine learning, data science, and classification algorithms.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron specifically detects uppercase abbreviations or acronyms (multi-letter all-caps tokens).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    In
    -1.45
    illige
    -1.45
    ensore
    -1.21
    ALLE
    -1.19
    DCC
    -1.16
    celles
    -1.16
     الاث
    -1.16
    {§
    -1.16
     [
    -1.15
    aient
    -1.14
    POSITIVE LOGITS
     between
    1.29
     για
    1.28
     Gutes
    1.21
    嫆
    1.21
     ему
    1.16
     verheiratet
    1.15
     jahr
    1.13
     Tools
    1.13
    )
    1.13
    荀
    1.12
    Activations Density 0.011%

    No Known Activations