© 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. 104477
    Prev
    Next
    INDEX
    Explanations

    connecting items with 'and' or '/'

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

    The neuron strongly activates on the key subject‐words that label or introduce a section’s main topic (e.g. “networking,” “risk,” “timer,” “planning,” etc.).

    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
     etc
    -1.04
     evtl
    -0.96
     поскольку
    -0.96
    Ǿ
    -0.94
     poiché
    -0.94
     ponieważ
    -0.93
    igeon
    -0.91
    などが
    -0.90
     dece
    -0.88
    圻
    -0.87
    POSITIVE LOGITS
     on
    0.94
     kering
    0.94
    рек
    0.94
     related
    0.93
    Rptr
    0.91
    érrez
    0.88
     bouch
    0.87
     brugt
    0.86
     in
    0.86
    viembre
    0.85
    Activations Density 0.104%

    No Known Activations