© 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. 53871
    Prev
    Next
    INDEX
    Explanations

    acrylic, plexiglass, PMMA

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

    The neuron detects mentions of technical polymer or chemical compound names (e.g., “acrylic,” “methacrylate,” “acid,” etc.).

    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
    festival
    -0.77
    aar
    -0.76
     ceramic
    -0.76
     Porcelain
    -0.75
    惱
    -0.73
    FEST
    -0.73
     Expectation
    -0.72
    鞭
    -0.72
    ウッド
    -0.71
    缕
    -0.71
    POSITIVE LOGITS
     acrylic
    1.41
     アクリル
    1.16
     Plex
    1.13
    Acrylic
    1.13
     Clear
    1.05
     Acrylic
    1.05
    透明
    1.03
     clear
    1.02
     plexig
    0.98
     PM
    0.91
    Activations Density 0.034%

    No Known Activations