© 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. 65783
    Prev
    Next
    INDEX
    Explanations

    semiconductor materials and devices

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

    The neuron strongly fires on numeric and punctuation tokens used in equations—especially parentheses and numbers—i.e. inline mathematical or formula snippets.

    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
     préciser
    -0.86
    冽
    -0.83
    îte
    -0.82
    цца
    -0.81
     Apocalypse
    -0.81
    ampunk
    -0.79
     Guangzhou
    -0.79
     pouvait
    -0.79
    igshid
    -0.78
    #+#
    -0.77
    POSITIVE LOGITS
     Semiconductor
    1.27
     semiconductor
    1.25
     doping
    1.13
     wafer
    1.12
    iconductors
    1.03
     крем
    1.01
     Silicon
    1.00
     carriers
    0.98
     epit
    0.97
    silicon
    0.97
    Activations Density 0.068%

    No Known Activations