© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27383
    Prev
    Next
    INDEX
    Explanations

    image references

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

    The neuron preferentially fires on longer or more complex tokens—especially multi‐character words, markup syntax, and file‐name fragments.

    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
     Grimal
    -0.93
    Auszeichnungen
    -0.92
     Chỉ
    -0.92
     Tables
    -0.90
    ึกษา
    -0.88
    テンツ
    -0.86
     elevated
    -0.85
    uidado
    -0.84
     Traducción
    -0.84
    ńskie
    -0.84
    POSITIVE LOGITS
     images
    1.38
     pictures
    1.11
     image
    1.08
     photos
    0.96
    images
    0.95
     picture
    0.94
    ◊
    0.93
     reizen
    0.85
    Obr
    0.84
     photo
    0.84
    Activations Density 0.094%

    No Known Activations