© 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. 23595
    Prev
    Next
    INDEX
    Explanations

    with numbers and chemical terms

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

    technical or scientific writing in academic/medical papers, particularly focusing on methodology sections.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on numeric tokens and numerical expressions (e.g. counts, measurements, percentages) in the text.

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

    technical, research-abstract style biomedical/scientific passages, especially those detailing experimental methods, measurements, units, and statistical results

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    scientific or technical text, particularly in academic research papers and medical/biological studies.

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
     outlining
    -1.03
     simplistic
    -1.03
     perplexing
    -1.02
     plainly
    -1.01
    と思って
    -1.01
     invigor
    -0.99
    と思ったら
    -0.98
     disregarding
    -0.97
     disheart
    -0.97
     broadening
    -0.96
    POSITIVE LOGITS
    他知道
    1.09
    他想
    0.94
     fazem
    0.93
    usses
    0.92
    他能
    0.91
    ⼰
    0.85
    他没有
    0.85
    他要
    0.85
     {
    0.84
    nables
    0.84
    Activations Density 5.699%

    No Known Activations