© 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. 11253
    Prev
    Next
    INDEX
    Explanations

    vs

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

    This neuron fires strongly on tokens that occur around the 3,000-token mark in the context window, effectively marking when you’re near the end of a long input.

    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
    QUEST
    0.60
    𝚣
    0.59
     సంబంధ
    0.57
     Investigative
    0.53
    лога
    0.53
    crossed
    0.53
    ountains
    0.52
     दर्ज
    0.52
    мно
    0.52
    ْل
    0.52
    POSITIVE LOGITS
     '',
    0.62
     !=
    0.61
    =='
    0.60
     Eigenschaften
    0.60
    '=>'
    0.58
     >
    0.58
     \,,
    0.58
     itibaren
    0.58
    '",
    0.56
     ซึ่ง
    0.55
    Activations Density 0.000%

    No Known Activations