© 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. 31-GEMMASCOPE-2-RES-16K
    7. 15907
    Prev
    Next
    INDEX
    Explanations

    space characters

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

    This neuron is attuned to code‐style or diff markers and URL‐encoding syntax—e.g. lines beginning with "+" or containing "echo" in patches and "%xx" escape sequences.

    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
    Pi
    0.64
    pi
    0.57
    pt
    0.55
    Render
    0.55
    Cup
    0.55
    行
    0.53
    Inc
    0.52
    bit
    0.52
    inc
    0.52
    位
    0.52
    POSITIVE LOGITS
     space
    1.74
     Space
    1.70
     SPACE
    1.67
     spazio
    1.58
     espacio
    1.57
     espaço
    1.55
     공간
    1.55
    空间
    1.51
    SPACE
    1.51
    space
    1.49
    Activations Density 0.043%

    No Known Activations