© 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. 12926
    Prev
    Next
    INDEX
    Explanations

    code blocks and commands

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

    The neuron primarily activates on sentence-ending punctuation (the period token marking end of a sentence).

    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
    O
    0.62
    C
    0.60
    D
    0.59
    J
    0.57
    A
    0.56
     D
    0.55
    E
    0.54
    G
    0.53
    N
    0.52
     C
    0.51
    POSITIVE LOGITS
     continú
    0.60
    与此同时
    0.60
     repercussions
    0.59
     нажа
    0.59
     unlink
    0.59
     мысли
    0.58
     しかし
    0.58
    কিন্তু
    0.57
     deleteAll
    0.57
     setTimeout
    0.56
    Activations Density 0.001%

    No Known Activations