© 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. 27399
    Prev
    Next
    INDEX
    Explanations

    interacting conditions and floor tiles

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

    The neuron highlights semantically rich content words—especially technical or domain‐specific nouns and verbs—while ignoring common function words.

    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
     varias
    -1.02
    putes
    -0.99
    诙
    -0.94
     todas
    -0.94
    рный
    -0.93
     двумя
    -0.91
    bergen
    -0.90
     mengatasi
    -0.88
     thinks
    -0.88
     Using
    -0.87
    POSITIVE LOGITS
    笑了笑
    1.05
     juf
    1.03
     organisme
    1.02
     maska
    1.00
    罍
    0.98
    …,
    0.98
     菇
    0.96
     kupa
    0.96
     皇族
    0.95
     zub
    0.95
    Activations Density 0.026%

    No Known Activations