© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38974
    Prev
    Next
    INDEX
    Explanations

    class labels and targets

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

    The neuron is detecting numeric tokens (e.g. integer or decimal literals) in code and text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ⓸
    -1.07
     $\&$
    -1.06
     ---------------
    -1.02
    
    -1.02
    евич
    -1.00
    邶
    -1.00
     $*$
    -0.97
    hassee
    -0.95
    managing
    -0.95
     Hauses
    -0.94
    POSITIVE LOGITS
     
    1.17
     labelling
    1.07
    лав
    1.01
    iações
    0.98
     labeling
    0.97
     Often
    0.91
     сигнал
    0.91
    IActionResult
    0.90
     Faça
    0.90
    label
    0.89
    Activations Density 0.012%

    No Known Activations