© 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. 22-GEMMASCOPE-RES-131K
    7. 26356
    Prev
    Next
    INDEX
    Explanations

    discussing conclusions and implications

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

    The neuron fires on language signaling the wrap‐up of scholarly arguments—words like “concludes,” “discussion,” “implications,” “recommendations,” and “results.”

    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
     är
    -1.14
     memenangkan
    -1.06
     reconocer
    -1.04
    ⌀
    -1.00
     kehilangan
    -0.98
     ersten
    -0.97
     获取
    -0.96
     pierws
    -0.96
     Document
    -0.96
     ditentukan
    -0.96
    POSITIVE LOGITS
     briefly
    1.05
     future
    1.04
    瞇
    1.02
     fleuve
    1.01
    ඨ
    1.00
     Schluss
    1.00
    olish
    1.00
     futurs
    0.99
     toekomst
    0.99
     hopefully
    0.99
    Activations Density 0.007%

    No Known Activations