© 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. 12441
    Prev
    Next
    INDEX
    Explanations

    not just

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

    The neuron fires on instances of the word “not,” marking negation or contrastive turns.

    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
    のですが
    0.98
    ่
    0.96
    Exist
    0.95
    Translate
    0.88
    क
    0.81
    usian
    0.80
    official
    0.78
     Conta
    0.78
    서
    0.77
    zeit
    0.77
    POSITIVE LOGITS
    ਾ
    0.97
     JHEP
    0.91
    钯
    0.89
    tingham
    0.86
    ول
    0.85
    larım
    0.84
    را
    0.82
     devons
    0.82
    lng
    0.81
    ولات
    0.81
    Activations Density 0.081%

    No Known Activations