© 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. 68411
    Prev
    Next
    INDEX
    Explanations

    comparison and contrast

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

    The neuron fires on subword pieces that make up hyphenated or compound technical terms (e.g. “non-”, “read-only”, “sur”, “ge-”, “currents”).

    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
     动物
    -1.03
     finalidad
    -0.89
     EconPapers
    -0.88
     koszulka
    -0.88
    firmasi
    -0.88
     rừng
    -0.88
    YNAMICS
    -0.87
    ujemy
    -0.86
     advertencia
    -0.86
    nonumber
    -0.85
    POSITIVE LOGITS
     unlike
    2.08
     Unlike
    1.52
    unlike
    1.47
    Unlike
    1.35
    不像
    1.28
     compared
    1.22
     puisqu
    1.11
     can
    1.10
     whereas
    1.08
     almost
    1.00
    Activations Density 0.080%

    No Known Activations