© 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. 31-GEMMASCOPE-2-RES-16K
    7. 2940
    Prev
    Next
    INDEX
    Explanations

    JSON structure and commands

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

    The neuron is specialized to pick out code line labels and markers—tokens like line‐number prefixes ending in “:”, or punctuation markers such as “;” and “);” that denote code statements.

    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
    ↵↵
    0.97
    ,:
    0.61
    )‏
    0.61
    ↵
    0.60
     $\}$
    0.60
     deverá
    0.59
     namun
    0.58
     jedoch
    0.58
     但是
    0.58
    ,":
    0.57
    POSITIVE LOGITS
    Finally
    0.61
    ້າງ
    0.59
    ﻓ
    0.55
    expanding
    0.53
    pro
    0.53
    Cl
    0.52
    Fair
    0.52
    magn
    0.52
    Novel
    0.52
    Pro
    0.52
    Activations Density 0.000%

    No Known Activations