© 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. 34-GEMMASCOPE-RES-131K
    7. 92964
    Prev
    Next
    INDEX
    Explanations

    process by reaction

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

    The neuron flags the technical, step-by-step “method” language you see in chemical patents—words like comprises, involves, contacting, reaction, produced, etc., that describe procedural process steps.

    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
     Только
    -0.81
    Exploration
    -0.77
    謨
    -0.75
    ikum
    -0.73
    Rach
    -0.72
    Buon
    -0.72
     manif
    -0.72
    Только
    -0.71
     раскры
    -0.70
    擬
    -0.70
    POSITIVE LOGITS
     heating
    1.91
     subjecting
    1.75
     exposure
    1.73
     exposing
    1.66
     Heating
    1.52
    exposure
    1.51
     heated
    1.49
     expose
    1.47
     subjected
    1.46
    Exposure
    1.44
    Activations Density 0.194%

    No Known Activations