© 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. 115728
    Prev
    Next
    INDEX
    Explanations

    chaperones and license strings

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

    The neuron fires most strongly on long runs of whitespace or indentation (e.g. deeply indented code or formatted markup).

    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
     and
    -0.98
    }{*}{}
    -0.98
    дизайн
    -0.93
     from
    -0.92
     senin
    -0.90
     também
    -0.87
     sosyal
    -0.85
    pierw
    -0.83
    »;
    -0.83
     envio
    -0.82
    POSITIVE LOGITS
     Dusche
    1.00
    }\
    0.98
     cúp
    0.96
    |
    0.96
     hordes
    0.94
    best
    0.93
     angep
    0.91
     debacle
    0.91
     .......
    0.91
     nécess
    0.88
    Activations Density 0.000%

    No Known Activations