© 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. 128471
    Prev
    Next
    INDEX
    Explanations

    numerical identifiers

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

    The neuron activates on numeric tokens and sequences (individual digits or multi‐digit numbers).

    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
     and
    -2.31
     Although
    -2.20
     but
    -2.20
     However
    -2.19
     In
    -2.13
     is
    -2.05
     For
    -2.03
     According
    -1.91
     Because
    -1.91
     as
    -1.91
    POSITIVE LOGITS
    鵙
    2.39
    厹
    2.36
    who
    2.14
    ꪹ
    2.11
    糴
    2.08
    睃
    2.08
    ⭓
    2.03
    ఽ
    2.02
    〬
    1.98
    ilfe
    1.95
    Activations Density 0.090%

    No Known Activations