© 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. 13772
    Prev
    Next
    INDEX
    Explanations

    technology companies | specific methodology | AI models

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

    This neuron consistently activates on the very first significant content token (the first non‐zero word) at the start of each new text segment.

    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
    t
    1.09
    le
    0.95
    ب
    0.91
    se
    0.91
    в
    0.84
    mn
    0.79
    men
    0.78
    m
    0.78
    6
    0.77
    e
    0.77
    POSITIVE LOGITS
     SmackDown
    0.98
     fermion
    0.93
     OVA
    0.91
     LHC
    0.90
     siguiente
    0.90
     planta
    0.88
     EPL
    0.87
     ESM
    0.87
     HRV
    0.85
     cisplatin
    0.85
    Activations Density 0.000%

    No Known Activations