© 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. 5688
    Prev
    Next
    INDEX
    Explanations

    video series, exception rule, gangster rap

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

    The neuron is detecting shell/terminal commands (e.g. “cd”) and other imperative command‐style tokens in technical how-to instructions.

    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
     संय
    0.80
     JobList
    0.79
    ،
    0.76
    Bart
    0.76
    Baz
    0.73
     टि
    0.73
    Birthday
    0.71
     ভয়ানক
    0.71
    asidd
    0.70
    ℍ
    0.70
    POSITIVE LOGITS
    ure
    0.98
     heme
    0.86
     veer
    0.86
    het
    0.81
    نګه
    0.80
    ä
    0.80
    еры
    0.79
    тная
    0.79
    opp
    0.78
    ेक्स
    0.78
    Activations Density 0.000%

    No Known Activations