© 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-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 14230
    Prev
    Next
    INDEX
    Explanations

    adverbs followed by adjectives

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

    The neuron detects adverbs ending in “-ly.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ly
    1.37
    ordinary
    1.03
    ar
    1.02
    ductory
    1.02
    ق
    0.98
    сть
    0.95
    fully
    0.95
    ness
    0.93
    ように
    0.93
    াস
    0.92
    POSITIVE LOGITS
     hablando
    1.02
     Lembrando
    0.84
    оз
    0.79
     falando
    0.79
    ﺕ
    0.79
     봤
    0.75
    ей
    0.75
     মিলিত
    0.75
     MDLVertex
    0.74
    ﻝ
    0.74
    Activations Density 0.291%

    No Known Activations