© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 84845
    Prev
    Next
    INDEX
    Explanations

    break

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

    The neuron responds to occurrences of “break” (in any inflected form such as break, breaks, broke, breaking).

    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
    也会
    -1.46
     podczas
    -1.45
     ſte
    -1.40
    鮰
    -1.39
    </h3>
    -1.39
     possibility
    -1.39
     ‚
    -1.38
     answers
    -1.37
     opportunities
    -1.31
    UGH
    -1.30
    POSITIVE LOGITS
    break
    1.92
    3
    1.79
     breaking
    1.78
     Breaking
    1.74
    Breaking
    1.70
    breaking
    1.66
     broken
    1.61
    2
    1.57
     Broke
    1.50
    broken
    1.48
    Activations Density 0.043%

    No Known Activations