© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 76163
    Prev
    Next
    INDEX
    Explanations

    reporting and aggregation

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

    The neuron primarily detects underscore characters—especially as part of snake_case identifier tokens in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     though
    -0.94
    liothèque
    -0.94
    ContentValues
    -0.80
    elemField
    -0.79
     esperado
    -0.77
     tráiler
    -0.77
     LU
    -0.76
     transaction
    -0.76
     Batalla
    -0.76
     TRA
    -0.75
    POSITIVE LOGITS
     Aggreg
    0.93
     aggreg
    0.87
     report
    0.84
    Aggreg
    0.84
     aggregates
    0.83
    报表
    0.82
     aggregation
    0.82
     absc
    0.82
    ração
    0.81
    ikon
    0.81
    Activations Density 0.001%

    No Known Activations