© 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. 111072
    Prev
    Next
    INDEX
    Explanations

    chemical formulas and structures

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

    chemical formulas and molecular structures in scientific papers.

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

    The neuron selectively activates on numeric tokens representing stoichiometric subscripts and other digits in chemical formulas.

    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
    ようになります
    -0.95
     Comunicación
    -0.93
    ようです
    -0.91
    影像
    -0.85
     accordingly
    -0.84
     allegedly
    -0.83
     Paesi
    -0.83
    A
    -0.83
    b
    -0.83
    wend
    -0.82
    POSITIVE LOGITS
    elekt
    0.96
     TAKEN
    0.93
    Lain
    0.91
    Geld
    0.90
     salonu
    0.90
     odeur
    0.89
    Kiri
    0.89
    Elekt
    0.88
     ramas
    0.88
    modellen
    0.88
    Activations Density 0.076%

    No Known Activations