© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 70401
    Prev
    Next
    INDEX
    Explanations

    code parsing and line numbering

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

    The neuron fires on numeric tokens that represent line or column numbers in structured code or error output.

    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
     eight
    -1.21
     so
    -1.20
     those
    -1.20
     six
    -1.09
    ！(
    -1.07
     four
    -1.06
     two
    -1.04
     seven
    -1.01
     nine
    -1.00
     what
    -1.00
    POSITIVE LOGITS
    癱
    1.12
     potenciales
    0.97
    Plusieurs
    0.94
    IVERSITY
    0.94
     billi
    0.94
     pivotal
    0.94
     serene
    0.93
    覗
    0.92
     EVEN
    0.92
     beset
    0.91
    Activations Density 0.001%

    No Known Activations