© 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. 10365
    Prev
    Next
    INDEX
    Explanations

    version check

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

    The neuron activates strongly on mentions of IP addresses, IP versions (e.g. “IP version 4,” “6”), and related networking configuration terms.

    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
    वण्यासाठी
    0.60
     subgoal
    0.57
     overland
    0.55
     removeFrom
    0.54
    飲み
    0.54
     binders
    0.54
    гот
    0.54
    眙
    0.52
     आत
    0.52
     পুনর্ব
    0.52
    POSITIVE LOGITS
     version
    1.92
    Version
    1.89
    version
    1.84
     Version
    1.84
     VERSION
    1.82
    版本
    1.75
    VERSION
    1.73
     versions
    1.71
    バージョン
    1.70
    getVersion
    1.70
    Activations Density 0.136%

    No Known Activations