© 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. 22-GEMMASCOPE-RES-131K
    7. 11100
    Prev
    Next
    INDEX
    Explanations

    integral part or mathematical integral

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

    The neuron fires on occurrences of the word “integral” (and related math‐integral notation) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    </b>
    -3.03
    </td>
    -2.48
     může
    -2.36
    ersoll
    -2.30
     avec
    -2.22
    ఽ
    -2.20
     esboço
    -2.19
    ፫
    -2.19
    不
    -2.17
    "{\
    -2.16
    POSITIVE LOGITS
    ization
    2.84
     Even
    2.52
     næste
    2.42
    </strong>
    2.41
     jajaja
    2.39
    T
    2.36
    !”
    2.23
    man
    2.19
    </h3>
    2.16
     Although
    2.13
    Activations Density 0.007%

    No Known Activations