© 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. 2243
    Prev
    Next
    INDEX
    Explanations

    numeric calculation

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

    This neuron reliably detects string literals (text enclosed in quotes) in code or code-like contexts.

    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
    ...).
    1.21
    …).
    1.14
    ?).
    1.10
    ...),
    1.08
    .);
    1.06
    ?),
    1.06
    ).</
    1.05
    !).
    0.98
    )».
    0.98
    .),
    0.92
    POSITIVE LOGITS
    ↵↵
    1.64
    ↵
    1.60
    ↵↵↵
    1.31
    ;
    1.30
     //
    1.26
                      
    1.22
     ||
    1.20
                            
    1.19
                       
    1.17
                                   
    1.17
    Activations Density 0.004%

    No Known Activations