© 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. 7878
    Prev
    Next
    INDEX
    Explanations

    breakdown of explanations

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

    This neuron fires on words expressing comprehension (e.g. “understand,” “understanding”).

    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
    ش
    0.93
    スク
    0.91
    IANS
    0.90
    RA
    0.87
    RO
    0.86
    ORA
    0.83
    ESE
    0.82
    UNG
    0.82
    ﻴ
    0.82
    IE
    0.81
    POSITIVE LOGITS
     how
    1.58
     why
    1.41
    how
    1.28
     cómo
    1.19
    why
    1.18
     warum
    1.15
    是如何
    1.09
     bagaimana
    1.06
     WHY
    1.06
     hvordan
    1.05
    Activations Density 0.910%

    No Known Activations