© 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. 84299
    Prev
    Next
    INDEX
    Explanations

    purification methods

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

    descriptions of scientific laboratory procedures for isolating and purifying substances.

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

    The neuron is picking out and strongly activating on descriptions of experimental procedures—especially purification and chromatography steps—in scientific method sections.

    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
     Mexican
    -0.85
     Accuracy
    -0.78
     kleinen
    -0.76
     accuracy
    -0.76
     محمد
    -0.74
     dziś
    -0.73
     precisión
    -0.73
    椿
    -0.72
    commen
    -0.72
    訪問
    -0.71
    POSITIVE LOGITS
     purification
    1.17
     dialysis
    1.13
     Purification
    0.96
    Purification
    0.95
    Шаг
    0.95
     purify
    0.93
     purifying
    0.92
     Fractions
    0.90
    Steps
    0.88
     fractionation
    0.88
    Activations Density 0.122%

    No Known Activations