© 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. 10-GEMMASCOPE-RES-131K
    4. 110746
    Prev
    Next
    INDEX
    Explanations

    really + intensifying phrase

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

    The neuron is detecting degree-intensifying adverbs (especially “really”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    "
    -1.94
    4
    -1.79
    9
    -1.59
     malgré
    -1.58
     a
    -1.55
    c
    -1.55
    遢
    -1.47
    >
    -1.45
     mere
    -1.40
     reason
    -1.39
    POSITIVE LOGITS
    sữa
    1.72
    giày
    1.59
     největ
    1.54
    áciu
    1.48
    thiết
    1.45
    Industrie
    1.44
     waarschijnlijk
    1.44
    máy
    1.42
     []
    
    1.40
    šní
    1.39
    Activations Density 0.015%

    No Known Activations