© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 11051
    Prev
    Next
    INDEX
    Explanations

    `open("..." as f`

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

    The neuron responds to programming API identifiers and function names (especially uppercase library acronyms like OGGZ and related I/O calls).

    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
     write
    0.64
    寫
    0.61
     écrire
    0.60
     Escrit
    0.59
    write
    0.58
     写
    0.57
     hard
    0.56
    写
    0.56
     Write
    0.55
    兒子
    0.54
    POSITIVE LOGITS
    opened
    1.31
     opened
    1.28
     Opened
    1.19
    が開
    1.13
     fopen
    1.11
    Opened
    1.10
    打开
    1.09
     відкри
    1.06
     Öff
    1.02
     opening
    1.01
    Activations Density 0.006%

    No Known Activations