© 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. 45743
    Prev
    Next
    INDEX
    Explanations

    drinking glasses and glassware

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

    The neuron is strongly triggered by mentions of drinking vessels—especially “glass” or “glasses” (e.g. beer glass, wine glass, champagne glass, glassware).

    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
    createConnection
    -0.82
     axios
    -0.79
     なっ
    -0.72
    SRR
    -0.71
    فته
    -0.67
    gemm
    -0.67
    itone
    -0.66
     tiendas
    -0.66
    dressing
    -0.65
    殷
    -0.65
    POSITIVE LOGITS
     glasses
    2.91
     glass
    2.67
    glasses
    2.05
     glassware
    2.03
    glass
    2.03
     Glasses
    2.00
     cup
    2.00
     Glass
    1.93
    グラス
    1.90
    Glass
    1.89
    Activations Density 0.018%

    No Known Activations