© 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. 33126
    Prev
    Next
    INDEX
    Explanations

    academic paper references

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

    The neuron is tuned to recognize standalone numeric tokens (e.g. years, volume/page numbers, and other digits) typically found in academic citations and references.

    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
    FAULT
    -0.97
     Why
    -0.94
    Novel
    -0.94
     auf
    -0.94
     since
    -0.94
    茄子
    -0.89
     BnF
    -0.89
    bigr
    -0.87
     Despite
    -0.87
    Bigr
    -0.87
    POSITIVE LOGITS
    身边
    1.05
     přid
    0.91
     lessened
    0.91
     quelles
    0.91
     recarg
    0.90
     HDD
    0.89
     캐릭터
    0.88
     daardoor
    0.88
     Lothar
    0.87
     puertos
    0.85
    Activations Density 0.005%

    No Known Activations