© 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. 75249
    Prev
    Next
    INDEX
    Explanations

    storing data and files

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

    The neuron strongly activates on terms and phrases related to storing or persisting data (e.g. “storing,” “store,” “data,” “string,” “blob,” “token,” “cookies,” etc.) in technical programming contexts.

    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
    lň
    -0.91
     first
    -0.84
     running
    -0.83
     stood
    -0.83
     to
    -0.83
    érez
    -0.82
    Di
    -0.82
    During
    -0.81
     प्राप्त
    -0.80
     Prime
    -0.79
    POSITIVE LOGITS
    ไว้
    1.08
    urée
    0.98
     indefinitely
    0.89
     potenci
    0.88
    chó
    0.87
    лока
    0.87
    enken
    0.86
     assur
    0.84
     jawaban
    0.82
    ]=(
    0.82
    Activations Density 0.132%

    No Known Activations