© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13145
    Prev
    Next
    INDEX
    Explanations

    Multiple choice options (D, E)

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

    The neuron strongly activates on the labeled answer choice (e.g. “A,” “B,” “C,” or “D”) in multiple‐choice question–answer sections.

    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_31_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
     TRUE
    0.75
     true
    0.71
     third
    0.71
     ●
    0.70
     ③
    0.70
     FALSE
    0.68
     ○
    0.67
     ②
    0.66
     al
    0.65
     thirds
    0.65
    POSITIVE LOGITS
    none
    0.98
    None
    0.93
     none
    0.83
     None
    0.74
    NONE
    0.74
    zuoye
    0.65
    goài
    0.63
    èves
    0.61
     NONE
    0.61
    сторон
    0.61
    Activations Density 0.004%

    No Known Activations