© 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. 34-GEMMASCOPE-RES-131K
    7. 12187
    Prev
    Next
    INDEX
    Explanations

    proper nouns or abbreviations

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

    This neuron fires on short code identifiers—especially the small lowercase letter clusters you see in Java package and class names (e.g. “mm”, “nq”, “sdk”, “ac”, 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_34/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
    どういう
    -0.91
     stressing
    -0.90
     affords
    -0.88
    しつつ
    -0.87
     stage
    -0.84
     stressed
    -0.83
     without
    -0.83
     χωρίς
    -0.82
     certain
    -0.82
     Verständnis
    -0.81
    POSITIVE LOGITS
     Kalifor
    1.07
     Litu
    1.06
     citroen
    1.00
     Godt
    1.00
     romantis
    0.98
    RTLU
    0.98
     kombi
    0.97
     kado
    0.96
    »).
    0.96
    huo
    0.94
    Activations Density 0.001%

    No Known Activations