© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13920
    Prev
    Next
    INDEX
    Explanations

    command to start

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

    The neuron detects promotional or spam‐style text—especially boilerplate marketing phrases and URL/path tokens.

    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
    ﺐ
    0.82
    Tere
    0.78
    باً
    0.77
    Everyone
    0.76
    קה
    0.76
    JT
    0.75
    erea
    0.74
    haft
    0.74
    ਣ
    0.74
    пан
    0.73
    POSITIVE LOGITS
     asociados
    1.03
     dúvidas
    0.96
     Ainda
    0.95
     alterações
    0.93
     conhecimentos
    0.92
     incluidas
    0.91
     informações
    0.90
     desafios
    0.90
     mudanças
    0.89
     asociada
    0.89
    Activations Density 0.000%

    No Known Activations