© 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. 12895
    Prev
    Next
    INDEX
    Explanations

    European language verbs/adverbs

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

    This neuron fires strongly on corporate‐style proper nouns and business entity names (e.g. company names, branded products, Inc., etc.).

    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
    ﺠ
    0.98
    ﺘ
    0.95
    ﻤ
    0.90
    ﺗ
    0.88
    ﻄ
    0.88
    ды
    0.86
    tenths
    0.86
    ﻘ
    0.85
    魈
    0.84
    teen
    0.83
    POSITIVE LOGITS
     voglio
    0.88
     prendere
    0.85
     bestimm
    0.83
     sugger
    0.82
     après
    0.80
     quando
    0.77
     cheveux
    0.76
     chiam
    0.75
    Deve
    0.75
     rencontre
    0.75
    Activations Density 0.000%

    No Known Activations