© 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. 14272
    Prev
    Next
    INDEX
    Explanations

    sequences of numbers started with 3

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

    The neuron consistently activates on the standalone token “3” (i.e. whenever the digit 3 appears as its own token).

    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
    ⟝
    -2.28
    )$
    -2.16
     übri
    -2.13
    所以我
    -2.06
    是
    -1.93
    不是
    -1.88
    我知道
    -1.82
    没有
    -1.82
    经常
    -1.79
    b
    -1.78
    POSITIVE LOGITS
    PUESTA
    2.58
    because
    2.28
     MUNICIP
    2.11
    躋
    1.86
    𖡼
    1.80
     administra
    1.77
    擄
    1.76
     orientale
    1.76
    There
    1.73
     renang
    1.73
    Activations Density 0.010%

    No Known Activations