© 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. 28271
    Prev
    Next
    INDEX
    Explanations

    search by field

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

    This neuron detects programming identifiers composed of capitalized words or acronyms (e.g., class names, method names, or other code tokens).

    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
    programme
    -1.07
     imprescin
    -1.00
    obviously
    -0.98
    −/−
    -0.98
     amazingly
    -0.96
     membro
    -0.96
     ludicrous
    -0.96
     joyful
    -0.94
     astonishing
    -0.94
     serene
    -0.93
    POSITIVE LOGITS
     and
    1.35
     or
    1.14
     répondu
    1.02
     gånger
    1.01
     by
    1.00
    eqref
    1.00
     (
    0.98
     åren
    0.97
     février
    0.97
    אם
    0.96
    Activations Density 0.007%

    No Known Activations