© 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. 103848
    Prev
    Next
    INDEX
    Explanations

    repeated symbol sequences

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

    The neuron strongly activates on long runs of the same punctuation character (e.g. “################” or “////////////////”), i.e. section-divider lines made of repeated symbols.

    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
    ---------------
    -1.02
     this
    -1.00
     all
    -0.97
     by
    -0.96
    ----------------
    -0.92
    ==============
    -0.90
    ขอบคุณ
    -0.88
     entusiasmo
    -0.88
     undoubtedly
    -0.85
     out
    -0.85
    POSITIVE LOGITS
    %%
    1.11
    Ahh
    1.07
    何も
    1.05
    %%%
    1.00
     beijo
    0.99
    imental
    0.98
     😀
    0.97
    身后的
    0.96
     começo
    0.96
     Willi
    0.95
    Activations Density 0.027%

    No Known Activations