© 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. 6012
    Prev
    Next
    INDEX
    Explanations

    This neuron detects string‐building or concatenation operations (e.g. uses of “+=”, “+”, or similar concatenation in code).

    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
     stejně
    0.71
     "=",
    0.67
     nama
    0.66
    一样
    0.66
    同様
    0.63
     '=',
    0.63
     LastName
    0.63
     "/",
    0.62
     similarly
    0.61
     nameof
    0.61
    POSITIVE LOGITS
    ('''
    1.17
    ("""
    1.16
    '''
    1.01
     '''
    0.99
    """
    0.99
     """
    0.88
    ```
    0.75
    .'''
    0.69
    ."""
    0.68
    """.
    0.68
    Activations Density 0.008%

    No Known Activations