© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 59438
    Prev
    Next
    INDEX
    Explanations

    punctuation marks

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

    The neuron responds to clusters of punctuation characters, i.e. multi‐symbol punctuation sequences like “.,”, “.:”, “.;”, “.-”, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     or
    -2.64
     and
    -2.31
     from
    -2.20
    で
    -2.16
     sooooo
    -2.08
     of
    -1.80
    cachor
    -1.79
     What
    -1.74
    戆
    -1.73
    。
    -1.71
    POSITIVE LOGITS
    they
    2.41
    其妙
    2.20
    ーーー
    2.13
    鹩
    2.13
    僂
    2.11
    卨
    2.05
    selben
    2.03
     juſ
    2.00
    蓥
    2.00
    鍆
    1.98
    Activations Density 0.027%

    No Known Activations