© 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. 10291
    Prev
    Next
    INDEX
    Explanations

    symbols before specific terms

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

    The neuron primarily activates on file‐system path fragments (especially Windows‐style paths with drive letters and backslashes).

    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
     
    0.58
    
    0.56
    
    0.51
    ?..
    0.50
     Nº
    0.50
    ".[
    0.50
     
    0.49
     
    0.49
    ?...
    0.49
    
    0.49
    POSITIVE LOGITS
    x
    0.44
    0
    0.43
    Sync
    0.43
    件
    0.42
    3
    0.41
    4
    0.41
    模
    0.41
    +
    0.40
    sync
    0.40
    recv
    0.40
    Activations Density 0.028%

    No Known Activations