© 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. 15744
    Prev
    Next
    INDEX
    Explanations

    same as

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

    This neuron fires on occurrences of the word “same,” especially when used to draw a comparison or indicate repetition.

    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
     Elektr
    1.13
     I
    1.09
    is
    1.02
     Lemmas
    1.02
     echi
    1.01
    royalblue
    1.01
     Ally
    1.00
     Agr
    0.98
     Mongol
    0.98
     Avg
    0.97
    POSITIVE LOGITS
    Same
    1.20
    轍
    1.19
    সঙ্গে
    1.15
    ように
    1.12
     সাথে
    1.10
     zamanda
    1.10
    ͝
    1.08
    同一
    1.06
    সাথে
    1.06
    同一个
    1.04
    Activations Density 0.033%

    No Known Activations