© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 107732
    Prev
    Next
    INDEX
    Explanations

    academic citations and version numbers

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

    The neuron fires on numeric tokens—especially the volume, issue, page, year, or reference numbers found in academic citations.

    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
     any
    -0.89
     blatantly
    -0.88
     on
    -0.87
     dozens
    -0.84
     your
    -0.84
     strives
    -0.83
     trying
    -0.77
     drastically
    -0.77
     with
    -0.76
    прода
    -0.76
    POSITIVE LOGITS
     SPECIAL
    1.01
     toujours
    0.96
    unsqueeze
    0.90
     monstru
    0.88
     PMID
    0.88
    SPECIAL
    0.88
     財布
    0.88
     monstrous
    0.87
    always
    0.84
    Segu
    0.83
    Activations Density 0.002%

    No Known Activations