© 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. 22-GEMMASCOPE-RES-131K
    7. 15897
    Prev
    Next
    INDEX
    Explanations

    distance, size, plates

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

    The neuron specializes in picking out numerical tokens and measurements (digits, numeric values, and related punctuation) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     igre
    -1.02
    werden
    -0.99
    =?";
    -0.93
    遙
    -0.90
     materie
    -0.89
    obras
    -0.88
     bude
    -0.88
     puse
    -0.87
    饉
    -0.86
    fundo
    -0.84
    POSITIVE LOGITS
     location
    0.95
     įsi
    0.94
    ֶר
    0.92
     vooral
    0.91
     unwarran
    0.89
    чника
    0.85
     henvis
    0.82
    НИЙ
    0.81
     Moderato
    0.80
     ngờ
    0.79
    Activations Density 0.150%

    No Known Activations