© 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. 107582
    Prev
    Next
    INDEX
    Explanations

    URL routing and config files

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

    The neuron fires on programming‐style identifiers and keywords (e.g. property names, variable or constant names, config keys) rather than ordinary natural‐language 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_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
    早めに
    -0.90
     modelling
    -0.86
    laining
    -0.84
     zape
    -0.81
     посвящен
    -0.81
     Mere
    -0.79
    örté
    -0.79
    wenye
    -0.78
    centre
    -0.77
    herself
    -0.77
    POSITIVE LOGITS
    full
    1.18
     full
    1.05
     alongside
    0.94
    Full
    0.92
    outState
    0.78
    FULL
    0.74
    Disclaimer
    0.72
     momentarily
    0.71
     besti
    0.71
    сл
    0.71
    Activations Density 0.012%

    No Known Activations