© 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. 24-GEMMASCOPE-2-RES-16K
    7. 1125
    Prev
    Next
    INDEX
    Explanations

    >

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

    This neuron is triggered by tokens that are part of code snippets—such as punctuation (e.g. “.”), uppercase identifiers, and scope‐resolution symbols (e.g. names ending in “::” or single uppercase letters like “D”).

    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
     harnessing
    0.60
     這個
    0.54
     这
    0.51
     這
    0.51
     эту
    0.51
     fontsize
    0.50
     pptd
    0.50
     abol
    0.50
     этой
    0.49
     glBindTexture
    0.49
    POSITIVE LOGITS
    create
    0.58
    get
    0.56
    Get
    0.53
    I
    0.52
    Create
    0.51
    V
    0.50
    []
    0.48
    R
    0.48
    P
    0.47
    Abstract
    0.47
    Activations Density 0.000%

    No Known Activations