© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 8627
    Prev
    Next
    INDEX
    Explanations

    emoticon punctuation

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

    The neuron is strongly detecting colon characters (":"), especially when used in punctuation or emoticons.

    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
     at
    -2.53
     from
    -2.42
     by
    -2.25
     as
    -2.11
     with
    -2.00
     istimewa
    -1.98
     While
    -1.91
     on
    -1.91
     in
    -1.88
     during
    -1.84
    POSITIVE LOGITS
     blatant
    2.13
    蓐
    1.91
     specif
    1.88
    臯
    1.86
    ︲
    1.81
    旁边的
    1.80
     ferocious
    1.80
     cuatro
    1.78
    苼
    1.77
    をつけ
    1.77
    Activations Density 0.021%

    No Known Activations