© 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. 118184
    Prev
    Next
    INDEX
    Explanations

    mathematics and symbolic expressions

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

    The neuron fires on uppercase letter sequences (e.g. the “R” in “ℝ” and the “NN” in “HNN”), i.e. it detects capital‐letter tokens or acronyms.

    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
     }_{\
    -1.01
    ᾽
    -0.97
     roleId
    -0.96
    𝛽
    -0.94
    }<
    -0.92
     comprob
    -0.91
     modelAndView
    -0.89
     изменя
    -0.87
    不得不说
    -0.86
    ]>=
    -0.85
    POSITIVE LOGITS
     regularly
    1.07
     hierbas
    1.04
     because
    1.03
     whenever
    1.00
    Baş
    0.99
    景点
    0.96
    labelControl
    0.96
    Recently
    0.95
    Numero
    0.94
     when
    0.93
    Activations Density 0.001%

    No Known Activations