© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9415
    Prev
    Next
    INDEX
    Explanations

    "{...}"

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

    The neuron activates strongly on substantive content words (specific nouns, verbs, and technical terms) rather than on common function words.

    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
    𝕌
    0.73
    خص
    0.73
    砲
    0.73
    Setter
    0.72
    ohan
    0.71
    Seine
    0.70
    這
    0.69
    Execution
    0.67
    Automation
    0.66
    Arsenal
    0.66
    POSITIVE LOGITS
     ܐ
    0.84
     glucocortic
    0.82
     Were
    0.79
     ହ
    0.79
     khắp
    0.79
     были
    0.78
     Stuck
    0.78
     ıs
    0.77
     Tert
    0.77
     Glut
    0.77
    Activations Density 0.001%

    No Known Activations