© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129226
    Prev
    Next
    INDEX
    Explanations

    The neuron detects domain-specific scientific acronyms and technical uppercase abbreviations (e.g. TNF, TBI, HSCT, PEG, TGF-β, Ixodes).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    after transforming growth factor

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

    commas following scientific or technical terms in academic or research contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
     dwóch
    -1.31
     dwie
    -1.13
     kilku
    -1.09
     två
    -1.09
     два
    -1.09
     δύο
    -1.06
     dwa
    -1.05
    two
    -1.05
    couple
    -1.04
    ทั้ง
    -1.04
    POSITIVE LOGITS
     one
    1.39
     during
    1.09
    راب
    0.97
     spurred
    0.95
     Each
    0.93
     amidst
    0.92
     as
    0.92
     each
    0.91
    óz
    0.91
     has
    0.90
    Activations Density 0.137%

    No Known Activations